University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange

Doctoral Dissertations Graduate School

6-1952

The Thermodynamics of and Its Compounds

James W. Cobble University of Tennessee, Knoxville

Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss

Part of the Chemistry Commons

Recommended Citation Cobble, James W., "The Thermodynamics of Technetium and Its Compounds. " PhD diss., University of Tennessee, 1952. https://trace.tennessee.edu/utk_graddiss/4376

This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:

I am submitting herewith a dissertation written by James W. Cobble entitled "The Thermodynamics of Technetium and Its Compounds." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Doctor of Philosophy, with a major in Chemistry.

William T. Smith, Jr., Major Professor

We have read this dissertation and recommend its acceptance:

Hilton A. Smith, Seymour Bernstein, G. E. Boyd

Accepted for the Council:

Carolyn R. Hodges

Vice Provost and Dean of the Graduate School

(Original signatures are on file with official studentecor r ds.) May 26, 1952

To the Graduate Council:

I am submitting to you a dissertation written by James w.1 Cobble entitled "The Thermodynamics of Technetium and Its Compounds. 1 I recommend that it be accepted in partial fulfillment of the require­ ments for the degree of Doctor of Philosophy, with a major in Chemistry.

We have read this dissertation and recommend its acceptance:

Accepted for the Council

,, THE THERMODYNAMICS OF TECHNErIUM AND ITS COMPOUNDS

A DISSERTATION

Submitted to The Graduate Council of The University of Tennessee in Partial Fulfillment of the Requ:irerents for the Degree of Doctor of Philosophy

by

James W. Cobble June, 19.52

,I ACKNOWLEDGEMENI'

The author wishes to express his appreciation to the Graduate School of the University of Tennessee, the Oak Ridge Institute of Nuclear Studies, and the Chemistry Division, Oak Ridge National Labora­ tory for their cooperation :in mak:ing possible the carrying out of this research through the Oak Ridge Resident Program. His deepest appreciation is expressed to Professor w. T. Smith, Jr., and Dr. G. E. Boyd, without whose earnest encouragement and guidance this dissertation would not have been possible. To

Margaret Ann TABLE OF CONTENTS

CHAPI'ER PAGE

Io INTRODUCTIONo • 00000000• . . . 1

II. PREPARATION AND IDENTIFICATION OF COMPOUNDS •. 4

Ao Preparation of Technetium o o o • 4

Bo Preparation of Technetium Heptoxide .. 7

C. Preparation of Pertechnic Acid o 9

D. Ammonium Pertechnetate •. ••• • 0 • • • • • • 9 E. Preparation of Technetium Dioxide • ••12 F. Technetium Trioxide ••••• . . 1 7 G. Other Compounds of Technetium • ...... • • 1 9 III. VAPOR PRESSURE STUDIES•••••• • • 21

A. The Vapor Pressure of Technetium Heptoxide••••• 21 B. The Vapor Pressure of Pertechnic Acid •• 32 c. The Vapor Pressure over Saturated Pertechnic Acid • • 35

IV• CALORIMETRY 0 0 0 0 . 0 . . 0 . . . . 39 A. The Heat of Combustion of Technetium Metal. . . . . 39 Bo The Heat of Solution of Technetium Hepto xide. . . . 53 c. The Heat of Dilution of a Pertechnic Acid Solution. . 0 . . . 0 . 0 ...... 56

v. OXIDATION REDUCTION CELLS 0 0 0 0 ...... 61

A. Acid Cell •• 0 0 0 000000000 62

B� Basic Cell. • 0 • • • • • • • • 67 V

C!fAPTER PAGE

VI. THERMOCHEMICAL CALCULATIONS • • • ••· ••...... 72 A. The Free Energies, Heats and Entropies of Formation of Technetium (VII) Compounds. . . 72 Bo The Free Energy, Heat and Entropy of Form9.tion of Technetium Dioxide ...... 78 c. The Free Energy, Heat and Entropy of Formation of Technetium Trioxide...... 79 D. The Oxidation-Reduction Scheme for

Technetium . . . . 0 . . . • • 0 ...... 82 E. Comparison of the Properties of Sub-

Group VII 0 ...... 82 VII. SUMMARY ...... 94

APPENDIXES. e>o••oo•••• . . . 96 I. X-RAY DIFFRACTION POWDER PATTERNS 97

II. PHYSICAL CONSTANTS, , .••••• 103 III. TREATMENI' OF THE CALORIMETRIC DATA...... 105 IV. THE HEATS OF COMBUSTION OF

AND RHENIUM TRIOXIDE•••••••• •••••••• 108 V. ESTIMATION OF ENTROPY VALUES FOR TECHNEI'IUM AND ITS COMPOUNDS. ll5 ., BIBLIOGRAPHY • • • • • • • • o o • o • • • • • 122 LIST OF TABLES

TABLE PAGE

I. Weight Loss Study on the Reduction of Ammonium Pertechnetate ••• ...... 6 II. Analysis of Technetium Dioxides • • . . . . . 15

III. Summary of Known Technetium Compounds • 0 0 0 0 • • • • 20

IV. The Vapor Pressure of Solid Technetium

Heptoxide • . . • . • • • • • . • • • • • 0 • • • • • • 27

V. The Vapor Pressure of Liquid Technetium

Heptoxide . . . . . • o • • • • • • • • • • • • • • • • 28 VI. Comparison of Thermodynamic Properties of

Technetium and Rhenium Heptoxides .••• . .•31

VII. The Dissociation Pressure· of Pertechnic Acid. · • • 33 VIII. The Vapor Pressure of Water ov er Saturated

Pertechnic Acid Solutions •••••••••••••••37 IX. Thermodynamic Data for Salvation Reactions of Technetium and Rhenium Heptoxides •••••••• • 38

x. Heat of Combustion of Paraffin Oil Accelerator •• • 46 XI. Benzoic Acid Combustion Calibration •• • • 48 XII. Heat of Combustion of Technetium Metal •• • • 50 XIII. Calibration of Solution Calorimeter •• . . . • • • 57 XIV. Heat of Solution of Technetium Heptoxide. • 58

xv. EMF of the Technetium Dioxide-Pertechnetate Electrode in Pertechnic Acid Solutions •••••••••65 vii

TABLE PAGE

XVI. EMF of the Technetium Diox:ide-Pertechnetate

Electrode in Basic Solution •• .•••••••••• 70 XVII. The Heat of Formation of

Pertechnetate o ••• 0 • 0 0 0 e e 77 XVIII. Summary of Thermodynamic Properties of Technetium

and Its Compounds .•••• . . • 0 • • 83 XIX. Comparison of Properties of Sub-Group VII . • • 87 XX. Comparison of X-ray Diffraction Patterns

for Technetium and Rhenium Heptoxides • • • • • • 98

XXI. Comparison of Technetium Metal X-ray Patterns••• ...... · · · · · · · • 99

XXII. Interplanar Distances for Technetium Dioxide Prepared by Pyrolysis of Ammonium Pertechnetate •••• ...... 101 XXIII. Physical Constants•• ...... 104 XXIV. Heat of Combustion of Rhenium Metal • • 110 xxv. The Heat of Combustion of Rhenium Trioxide•• • 112 XXVI. Summary of Entropies Estimated fo r Technetium and Some of Its Compounds •. • •••••.•••••121 LIST OF FIGURES

FIGURE PAGE

1. Potentiometric Titration of a 0.1 M. Pertechnic

Acid Solution. • • • • • • • • • • • . • • 0 • 10

2. Beer's Law Plot for Aqueous Ammonium Per- technetate Solutions . . . • • • 0 13 J. Sickle Gauge • ...... • ...... 22 4. Sickle Gauge (Photograph) ...... • . . . . . 23 5. Vapor Pressure Apparatus ...... 26 6. Vapor Pressure of Technetium Heptoxide . . . 30

7. Vapor Pressure of H20 over HTc04(s) and IITc04(sat.)· • 34 8. Combustion Calorineter •• ...... 40 9. Combustion Calorimeter • 41

10. Micro-Combustion Bomb •. . . . . 43 ll. Solution Calorimeter .. • • 54 12. Oxidation-Reduction Diagram for Technetium •. • . 85 lJ. Oxidation-Reduction Diagram for Sub-Group VII. . 91

14. X-ray Powder Pattern for Technetium Dioxide..••• . 102 15. Time-Temperature Curve for Calorimeter •••. • 106 CHAPTER I

INTRODUCTION

The recent1 large scale separation of technetium from the fission products at the Oak Ridge National Laboratory .has made possible an investigation of the chemistry of technetium-using much larger quantities than have previously been availableo Although 2 element 43 was reported by the Noddacks in 1925 , they have not pub­ lished any further information on the element. In 19353 Perrier and SegrJ bombarded in a cyclotron with energetic deuteron particles and produced an isotope whose chemistry followed that expected for element 4Jo Because of the criticism of the Noddacks'

4 evidence of element 43 , ani the further fruitless attempts of others to repeat their work.5, 6, it now seems well established that the first bona fide chemical evidence for element 43 was that of Perrier and Segre. The International Union of' Chemistry has accepted their claim and the name which they gave to it, technetium. ?

11 (1) Performed by the Ho�' Laboratory Group, Chemistry Division, Oak Ridge National Laboratory. (2) Noddack, w., Tacke, I., and Berg, o., Naturwissenschaften, 13, 567 (1925). I (3) Perrier, c. and Segre, E., !!.:_Chem. Phys.,�' 712 (1937); ibid., 1, 155 (1939). (4) Prandtl, w., � angew. Chem., 39, 1049 (1926). (5) Prandtl, w., Ber., 60, 621 (1927). (6) von Hevesy, G., Chem. Rev., J, 321 (1927). (7) Committee on New Elements, International Union of Chemistry, Chem. Engo News, 27, 2996 (1939). 2

Until the occurrence of the chain reacting pile, only unweigh­ able quantities had been produced by cyclotron bombardments of molyb­ denum; much of the previous 11 tracer11 chemistry has been summarized recently by Hackneyo8 Since the War fractional milligram quantities9 have been produced by prolonged neutron irradiation of molybdenum in the Oak Ridge National Laboratory graphite reactor o Milligram quan­ titites were also isolated10 from fission product solutions from project sources in 1948. In the sunnner of 1951, fractional gram quantities became available for research purposes.11 It was felt that a thermodynamic study of -�he element and its compounds would be of primary interest in help:ing complete the thermo­ dynamic data available on the transition elements. Such data are of aid :in �derstanding the role of the d electrons in chemical bonding. In addition, the chemical similarities and differences of technetium and rhenium could best be demonstrated by quantitative data. Finally, such information should aid in predicting the stabiiity of, as yet, unknown technetium compounds o In this dissertation heats of solution, of oxidation and com­ bustion, of vaporization and sublimation, and free energies of reaction for some of the well identified valence states of technetium are reported.

(8) Hackney, J. c. , i!..:._ Chem. Ed., 28, 186 (1951).

(9) Motta, E. E., Boyd, G. E., and Larson, Q. v., Mon-c-169 (1946); Phys. Rev., 72, 1270 (1947). (10) Parker, G. w., Reed, J., and Ruch, J. w., AECD-2043 (1948). (11) _We are indebted to G. w. Parker and his group at the Oak Ridge National Laboratory for making available this element for the research presented in this dissertation. 3

·By use of some semi-theoretical considerations of entropy, both free energies and enthalpies of technetium compounds have been calculated from the experimental data. The stability, the redox scheme, and some of the bond strengths of technetium and its compounds have been calcu­ lated and compared with the information available for other members of Group VII. In general, the thermodynamic behavior of technetium and its compounds has been found to be that predicted from the position of the element in the periodic table. A few noteworthy ex­ ceptions, however, have been found. CHAPI'ER II

PREPARATION AND IDENTIFICATION OF COMPOUNDS

Since the previous work on technetium had been performed on quantities of a few milligrams or less, it was desirable to prepare larger amounts of the element and compounds, and to identify, charac­ terize, and prove their formulas by analysis. Technetium metal pro­ vides the best starting material, and it will be described first.

A. Preparation of Technetium Metal

The technetium was obtained from the 11Hot 11 laboratory at th e

Oak Ridge National Laboratory. It consisted of about lg. of technetium as tetraphenylarsonium pertechnetate with perchlorate carrier dissolved in 10 1. of concentrated sulfuric acid. In addition to the technetium , the crude material also contained considerable amounts of radioactive , although it was not present in quantities large enough to be considered chemically as but a trace :impirity • In addition, small quantities of rhenium were also present, since this material had been used by the 11 Hot 11 laboratory as a stand-in to test the separation procedure of technetium from the fission products.

It was found that the tetraphenylarsonium pertechnetate could be electrolyticalzy decomposed by electrolysis of liter quantities of the concentrated sulfuric acid solution with large surface area platinum electrodes to give a solid product. Subsequent distillation of the solid formed from sulfuric, perchloric and nitric acid mixtures 5 gave a distillate from which technetium sulfide could be precipitated. 1 The detailed procedure is described elsewhere, but briefly it con­ sisted of dissolving the sulfide in ammoniacal hydrogen to give ammonium pertechnetate and ammonium sulfate. This solid was then heated in a stream of hydrogen in a quartz tube. The reduction pro­ ceeded rapidly at low (200-3000) temperatures and the sample was finally raised to high enough temperatures (500-6000) to sublime the ammonium sulfate away from the technetium metal. The metal was de­ monstrated to have the same sin2e and intensity values as the X-ray 2 pattern taken by Mooney ' 3 on metal prepared by sulfide reduction.4 Further, the procedure could be used to reduce pure ammonium p ertech­ netate quantitatively to the metal as shown by data on reduction time as a function of loss in weight. (See Tabler.) Contrary to published results,5 the metal would not dissolve in hydrochloric acid of any concentration, hot or cold, nor would it react with ammoniacal hydrogen peroxide as does metallic rhenium. It did dissolve, however, in aqua regia and dilute or concentrated nitric

6 acid, in agreement with similar studies by Fried on smaller quantities 6

(1) Cobble, J. w., Nelson, C. M., Parker, G. w., Smith, W. T., Jr., and Boyd, G. E., !!.! Am. Chem. Soc., 74, 1852 (1952 ). (2) Mooney, Rose c. L., Acta Cryst., �, 161 (1948) . (3) For this data, see Appendix, Part I.

(4) Fried, s., �Am. Chem. Soc., 70, 442 (1948) .

(5) Hackney, J. ,C., �Chem. Ed., 28, 186 (1951). (6) Fried, s., op. cit. 6

TABLE I

WEIGHT LOSS STUDY ON THE HYDROGEN REDUCTION OF AMMONIUM PERTECHNETATE

wt. of Sample Heating Heating Wt. after Formula before Heating Time Temperature Heating for Residue

° 14.57 mg. 1 hr. 275 8.23 mg. 78.1% Tc02 or lOJ.3% Tc

8.2J 1 350 8.02 100.6% Tc 8.02 1 400 7.97 100.0 . 7 097 1 500 7.97 100.0 a 7. 97 1 900 7 .82 97.5 7.82 1 500 7.82 97.5

B.some loss of technetium metal by volatilization is indicated at this higher temperature. 7

The metal will also burn in to form technetium heptoxide,7 but reaction with proceeds very slowly, if at au.8 The pure, freshly reduced metal is silvery-white, similar in appearance to rhenium prepared by ammoniwn perrhenate reduction. It tarnishes in moist air to give a grey powder. The from X-ray data (assuming an atomic weight of 99.0) is 11.50 g./cc.9, and the is 2140°.10 The ionization potential of the metallic 11 gas is 7 .45 volts.

B. Preparation of Technetium Heptoxide

Technetium can be burned in dry oxygen at 40o-5ooo to give technetium heptoxide, Tc207, in an analogous manner to that used to prepare rhenium heptoxide. 12 The yellow solid formed melts at 120°, is extremely hygroscopic, and dissolves in water to give a solution of pertechnic acid, HTc04. The details of the proof of the formula

(7 ) Boyd, G. E., Gobble, J. w., Nelson, c. M., and Smith, W. T., Jr., � Am. Chem. Soc. , 7 4, 556 (19.52). (8) Nelson, c. M., 11The Magnetochemistry of Technetium and Rhenium", Doctoral Dissertation, The University of Tennessee, 1952. (9) Mooney, R. C. L. , Acta, Cryst. �, 161 (1948); Phys.�, 72, 1269 (1947). (10) Parker, G. w., and Martin, w. J., Oak Ridge National Labora­ tory, Chemistry Division Quarterly Report 1·or Period Ending December 31, 19.51, ORNL-1260.

(1 1) Meggers, W. F., � Research Na� Bur. Standards, 47, 7 (1951). (12) Melaveq, A. D., Fowle, J. N., Brickell, w. and Hiskey, C. F., "Inorganic Synthesis", Vol. III, McGraw Hill., New York, N, Y., 1950, p. 188. 8 of this have been previously described.13 The oxide, as well as rhenium heptoxide, was tested for its electrical conductivity by dis­ tilling it into a small glass bulb which contained two elec­ trodes and measuring the (D.C.) resistance. Rhenium heptoxide was found to conduct in the molten state, but technetium heptoxide apparently behaves quite differently. The solid at room temperature is non-conducting, but as the temperature is raised towards the melting point, the sample begins to shav conductance which increases with in­ creasing temperature until the melting point is reached. At its maximum conductance, it conducts about one-fourth as well as fused chloride in the same apparatus. When the roolting point is reached, however, the conductance rapidly drops to zero, and remains so up to the boiling point. This may apparently be some type of a premelting phenomenon; the melting point is rather low compared to rhenium heptoxide, although their boiling points are comparable. The melting process in technetium heptoxide is also accompanied by an abnormally large entropy change (See Chapter III). Technetium heptoxi.de is not isomorphous with rhenium heptoxide. (See Appendix L) Solutions of technetium heptoxide in concentrated sulfuric acid were observed to demonstrate various colors (red, green, blue) supposedly - due to ions such as Tc03fTco4 similar to those postulated for manga­ nese.14

(13) Boyd, G. E., 'Cobble, J. w., Nelson, C. M., and Smith, w. T. , Jr.,�� Chem. Soc. , 74, 556 (1952). (14) Latimer, w. M., and Hildebrand, J. H. , "Reference Book of Inorganic Chemistry", McMillan Company, New York, N. Y., 1943, p. 374. 9

c. Preparation of Pertechnic Acid

It has been sh,ownl5 that when an aqueous solution of technetiwn

heptoxide is slowly:�vaporated over a solution of sulfuric acid at room temperature, crystals are formed which analyze �s Tc207•H20 or HTc04. These crystals are reddish black in appearance16 and are deliquescent. If the dehydration is continued for a long enough time, the compound will loose its water and form the heptoxide. The acid is even more unstable in this respect than ; the vapor pressure char­ acterization of this compound will be discussed in Chapter III.

Aqueous solutions of pertechnic acid show all of the character­ istics of "strong" acids. Figure 1 is the potentiometric titration curve for a 0.1 M. solution of pertechnic acid with 0.5 M. hy­ droxide solution. From the inflection point at a pH of 6.5 ± 0.5, it may be concluded that the ionization constant of the acid is > 0.1.

n. Ammonium Pertechnetate

Ammonium pertechnetate remains from the evaporation of aqueous solutions of pertechnic acid to which an excess of ammonium hydroxide has been added. The of the compound has been deter-

(15) Boyd, G. E., Cobble, J. w., Nelson, C. M., and Smith, W. T., Jr.' ---J. Am. Chem. Soc., 74, 556 (1952). (16) Boyd, G. E., Cobble, J. w., Nelson, C. M., and Smith, W. T., Jr., ---loc. cit. 10

11. 0 --....-----.---.---.----.-----r---.--...--r--r----,,--�---r--, · ·-· --· 10.0 --·-·-·- - __ ..,'--�•' •. : / 1• 9.0 ,.I I ,f• 8.0 •

7.0

JiNFLECTION pH= .5 ±0.5 6.0 6

I C. •

5.0

• 4.0 3.0 ,, /I., •" I •• I • ...-•· I ...... I 2.0 ·-·-·-• ..-:::!�------, i...- 1. 0 .s-...... 1.1-o_ ...... 1_ -- ...L.2--. .,..__3 ..... 1 _a ----1._s--L---'-1 . --_'--2---J.3 _41--_...... _s--mis. 0.5M.-'-s-- NaOH.'--_ ...... 2.0 FIGURE 1.- POTENTIOMETRIC TITRATION OF A 0.1 M. PERTECHNIC ACID SOLUTION 11

17 mined: it is tetragonal and the unit cell contains 4 molecules. It is also isomorphous with ammonium perrhenate and annnonium periodate.

The lattice dimensions indicate that the perrhenate ion is larger th an the pertechnetate ion. Its density is 2. 73 g. cc.-1•

The salt is white when pure, can be dried in an oven at 95 -

105°, is non-deliquescent, and has proved a convenient gravimetric form in analyzing pertechnic acid solutions ani technetium heptoxide samples. Recently, ha.rever, Rulfs and Meinkel8 have questioned the stability of the salt under norm9.l drying-oven conditions . Since their observations, based on half-milligram quantities, were contrary to our previous experience, it seemed desirable to check them with larger samples. A 100 mgo sample was heated in a covered platinum dish at 80-100° for eight dayso After this time the sample had lost only 0.3 mg. or 0 o3 percent, having lost 0 o 26 percent,which was probably water, in the first 24 hourso From these and many other observations, we conclude that the pure salt is relativ�ly stable under these conditions. The same observations have been made by others on ammonium perrhenate.19

The pertechnetate ion has a strong u ltraviolet absorption spectrumo 20 This spectrum, which contains two strong peaks at 2470 A.

(17 ) Zachariasen, w. H., Argonne National Laboratory Report, ANL-4082, 1947; to be published in Acta. Cryst.

(18) Rulfs, c. L., and Meinke, w. w., .!!!_ Am. Chem. Soc., 74, 235 (1952).

(19) Smith, W. T., Jro, Private Communication, 1951.

(20) Boyd, G. Eo, Cobble, Jo W., Nelson, C. M., and Smith, W. T., Jr., i!_ Am . Chem. Soco, 74, 556 (1952). 12 and 2890 A., can be used for the quantitative determination of per­ technetate ion solutions. A Beer's law plot determined with a Beckman Model DU Spectrophotometer and fused silica cuvettes is shown in

Figure 2. Because of the hig h absorption, this zoothod can be used to determine extremel;y" small quantities of technetium.

E. Preparation of Technetium Dioxide

Fried21 has reported that a black compound, isomorphous with rhenium dioxide, is formed when ammonium pertechnetate is heated at

400-500° according to the following reaction:

(II-1)

The reaction must be carried out in a sealed tube to prevent subli­

1 mation loss of the ammonium salt. Because Fried s samples were so small, it was not possible for him to carry out a chemical identi­ fication of the substance, and his evidence is based largely upon

X-ray diffraction comparisons o Accordingly, the preparation has been attempted using somewhat larger samples (5-10 mg.) and analyzing them by conventional semi-micro methods. The black residue left after pyrolysis in a sealed quartz tube was dissolved in an excess of standard eerie sulfate solution and then the excess eerie was back-titrated with standard iodide. The dissolving of the oxide was rather slow, requiring about one-half hour.

(21) Fried, s. , Private Communication to G. E. Boyd, 1950. 2_5..------r------,------.------,------r-----,-----,-----,

0/4oA. 2.0

� 1.5 (f) z w 0 _J <{ 6 0 1 I- .0 0 • � 0 5 0. �

0/ � . / . 0 -� 0 o 5xl0-5 lxl0 -4 2xlo-4 3xlo-4 4xio-4 5xio-4 6xlo-4 7xlo-4 CONCENTRATION ( M.)

1 I-' FIGU RE 2.- BEERS LAW PLOT FOR AQU EOUS AMMONIUM PERTECHNETATE SOLUTIONS \.,.) 14 There is no apparent direct reaction between technetium dioxide pre­ pared in this manner and dilute iodide so lutions as is lmown in the case of . Table TI summarizes some analyses carried out on milligram amounts of the black compound prepared by pyrozysis. The molecular weight was calculated assuming that the oxidation by eerie ion is as follows:

(II-2)

This reaction has been found to hold for rheniwn dioxide samples. 22 The X-ray diffraction pattern of sample I (see Table II) was also taken and was found not to be the same as that reported by Fried, nor is it the same as that found for Tc02 hydrate prepared by solution reduction.23 This might be explained in terms of various high tempera­ ture modifications of the compound, or by the fact that. the products of the pyrolysis are not adequatezy stated by equation (II-1). It should be noted (see Table II) that the compounds formed have empiri­ cal formulas corresponding to either Tc02 or Tc02 hydrateo However, since they were all prepared under about the same con:iitions, there may be other formed also o The apparent agreement, then, to compounds of Tc02 or the hydrate nay be fortuitous, although the

X-ray diffraction pattern of Tc02 prepared in this manner is the same as the Tc02 prepared by heating Tc02 hydrate in a vacuum at high

(22) Geilman, w., and Wrigge, F. W., --z. anorgo u. allgem. -Chem., 222, 56 (1935).

( 23) Nelson, C. M., 11 The Magnetochemistry of Technetium and Rheniwn", Doctoral Dissertation, The University of Tennessee, 1952_. '-

TABLE II

ANALYSIS OF TECHNETIUM DIOXIDES

) Percent ·Nearest Run Conditions Sample- Wp o Moe. Wt. Mol. wt. Ca Nearest Formula Wt. Formula Wt.

I Sealed Tube 3.08 mg. 4.5.8 137 Tc02 : 131 103 200-300° (2 hr.) J.18 44.4 133 Average lJ;'

II Sealed Tube L73 .58.9 177 T c02 • 2If20 : 167 105 200-300° (2 hr.) 4.86 .57. 6 17:3 Average 175

III Evacuated Tube 3.20 5.5�2 166 Tc0 •2H 0 : 167 ° .so.o 2 2 103 200-300 (2 hr�) 3.55 L77 Average 172

: IV Sealed Tube 3.12 .52;6 158 Tc02•2H20 167 95 6000 (2 hr.) 3.35 60 180 4.40 4.5.9 138 Average 1.59

aAssuming 3 equivalents per mole of Tc(IV). � 16 temperatures, (See Appendix I ) •

The black na.terial so prepared can be titrated directly with eerie sulfate solution, although the titration proceeds very slowly. The reaction is:

(II-3)

From the inflection voltage of this titration, however, the oxidation

2 potential of equation (II-2) can be estimated from the formula: 4

0 0 aEred. + bEoxd. ' (II-4) a ;. b where E infl. is the inflection-voltage of the potentiometric titration, and a and b are the coefficients in equation (II-3) for the oxidizing 0 agent (eerie) an d reducing agent (Tc02). E stands for the red.-oxd. standard potentials of the ceric-cerous and the technetium dioxide - pertechnetate systems. Using 1.44 volts for the cerous-ceric potentia125 and the observed inflection voltage of 940 ± 100 millivolts, we calculate the standard potential for the Tc02 -Tco4- electrode to be -0.6 volts. The limit of error in this estimate is about f 0. 2 volts (or larger).

The largest uncertainty is the solubility of the technetium dioxide.

It will be seen, however, that this voltage is in agreement with the standard potential of this electrode system as measured directly. (See

(24) Kolthoff, I. M., and Furman, N. H., 11Potentiometric Titrations11 , John Wiley and Sons, New York, N. Y., 1931, p. 45. Kolthoff, I. M., and Furman, N. H., op. cit., p. 280. (25) 17 Chapter V).

26 Tc02 hydrate has also been prepared by Nelson from the solu­ tion reduction of �ertechnetate solutions with zinc an:i hydrochloric acid. It has been characterized by him and was also used in this re­ search. The X-ray diffraction pattern has also been taken for this compound, but it was similar neither to the pattern given by sample I (see Table II) prepared by pyrolysis in this research nor to that obtained by Fried.

It has been found that Tc02 can be plated out on a platinum electrode (cathode) from neutral or basic solutions. While the black material so formed has not been identified chemical.1¥, it gi.ves the same potential for the dioxide-perteohnetate electrode as the Tc02 prepared by Nelson. We can conclude, then, that the material plated out must be pure Tc02, or that any other products which may have been formed did not affect the voltage of the cell. The details of this cell identification appear in Chapter v.

F. Technetium Trioxide

During the preparation of technetium heptoxide a red material is frequently fonned in the combustion tube which behaves somewhat differently than the red solutions of concentrated pertechnic acid. This later solution becomes colorless on dilution, while the red material formed above when dissolved in water to give a vezy dilute

(26) Nelson, C. M., 11The Magnetochemistry of Technetium and Rhenium", Doctoral Dissertation, The University of Tennessee, 19.52. 18 colored solution (yellowish) apparently decomposes to give a black residue and pertechnetate ion. The black residue is technetium dioxide. The same behavior has been noticed with the red material obtained when organic vapors are allowed to come in contact with technetium heptoxide. Since rhenium trioxide is red, it may be that the red mterial so obtained is technetium trioxide. If this is true, then this compound is more like the �6 valence state of manganese than of rhenium. The probable reaction between water and technetium trioxide can be written stepwise: - Tc0 (TI-5) TcOJ(s) + H 20(1)-+ aq. : 2H!aq.) f 4(aq.) - + ( 6) 3Tc04(aq.) + 4H(aq.) = 2Tc04(aq.) f Tc02(s) f 2H2o(l) II-

Rhenium trioxide is insoluble in water. Attempts to prepare larger quantities of the trioxide using methods published by others27, 28 far rhenium trioxide have not been successful. Part of this difficulty is due to the fact that apparently the trioxide reacts with even trace quantities of water. It may also be that the trioxide is too unstable to allow accumulation of any but trace quantities of this material . A more thorough attempt to pre­ pare this compound has not been made, however, and its characterization is still incomplete.

(27) Nechamkin, H., Kurtz, A. N., and Hiskey, c. F., � Am. Chem. Soc., 73, 2828 (1951). (28) Melaven, A. D., Fowle, J. N. , Brickell, w., and Hiskey, C. F., "Inorganic Synthesis", Vol. III, McGraw-Hill, New York, N. Y., 19�, p. 187. 19 G. other Compounds of Technetium

Table III SWTllllarizes sol'm:l of the other compounds which have been reported for technetium but which were not prepared in connection with this research. In general, those of Nelson and Parker involved hundred milligram quantities, while the others were characterized wit h quanti­

ties of a few milligrams or less . TABLE III

SUMMARY OF KNOWN TECHNEI'IUM COMPOUNDS

Compound Investigator Color Properties

a Tc This Rese arch; Fried grey Inert to HCl; . oxi d. by aqu a regia 3 nitric acid.; . burns in o2 • NH4Tc04 Fried white Density : 2.73; tetr agon al; very soluble; relatively st able; su9l:µnes at � 1509 . b ° ° KTc04 Fried; Parker white Ks .p . = 0.44 at 27 , 0,14 �t 7 . AgTc04 Frieda Identified by X -r ay . diffr action. c K2TcCl6 Nelson yellow Hydrolyses in water; isomorphous with but larger cell than K2ReCl6. Tc02 Nelson; This Rese arch black. Inert; dissolves. in str ong oxid. agents; isomorphous with Re02 . Tc 207 This Rese arch yellow Delitjues cent; stable. up -to. B: . P. (3100);' dissolves readily in water to .give pertechnic acid; not isomorphous with Re2o7 • . HTc04 This Research red Strong aci d in water ; loses w ater at elevated temper atures; deliquescent. d Tc2S7 Rulfs and Meinke bl ack Not precipitated in very cone. acid solns.; c arries excess . a TcS2 Fried black Pr ep ared by ne ating Tc2S7 with excess sulfur; identified by X-ray -diffraction. . . . -· - -· . Tc03 This Research red Disproportionates in water ; not definitely char acterized aFried; - S. ; Private communication to ·G. ·E.- Boyd, 1950.- bparker 3 G. W., Oak Ridge National Labor atory Chemistry Division . Qu arterly Report,· 0RNL.,.1116, 1951. CNelson, c. M. , "The Magnetochemistry of Technetium an d Rhenium" , Doctor al Dissert ation, The University of· "' Tennessee, 1952. 0 dRulfs, C. L. , an d Meinke , w. w., �Am. Chem. Soc. , 74, 235 (1952 ). CHAPI'ER III

VAPOR PRESSURE STUDIES

A. The Vapor Pressure of Technetium Heptoxide

Technetium heptoxide was found· in course of preparation to be volatile. It was of interest, therefore, to study its vapor pressure as a part of the characterization of this compound. Similar measure­ ments have also been made on rhenium hepto:xide1 although manganese heptoxi.de is apparently too unstable to study in this manner.2 Since loss of technetium had oc curred in previous tracer experinents3 when various solutions of technetium had been evaporated to dryress, it was of interest to see if this loss could be accounted for by the volatility of the hepto:xide. This compound results on dehydrating acidic solutions of Tc(VII). Technetium heptoxide, like rhenium heptoxide, is corrosive and is reduced by both and organic vapors. An all-glass "sickle"

or Bourdon gauge was therefore used (see Figures 3 arxi 4 ) far the rooasurements. The gauge was designed after one described by Phipps4,

(1) Smith, W. T., Jr. , Line L., and Bell, W. A., -J. -Am. Chem. Soc., in press (1952). ( 2) Mellor, J. W., "Treatise on Inorganic and Theoretic al Chemistry", Vol. 12, Longmans, Green and Co., New York, N.Y., 1932, p. 466. (3) Haclmey, J. C., !!.:_ Chem. Ed., 28, 186 ( 1951) . (4) Phipps, T. E., Spealman, M. L., and Cooke, T. G., J. Chem. Ed., 12, 321 (1935). 22

I I I I I I I I I I I I I I I I I I I I I I l \"--- ____----___,

FIGURE 3.- SICKLE GAUGE 23

FIGURE 4.- SICKLE GAUGE 24

and similar to the one used by Smith.5 It was used as a differential or as a null-point indicator for the vapor pr essure of the liquid, and as an absolute manometer for the measurements on the solid hep­ toxide. The deflection was linear over a 2-3 mm. range, and it s sensi­ tivity was 1 mm o displacement/mmo pressure differential . A travel- in g microscope was used to read the relative displacement between a fixed, reference pointer and the index pointer. When used as a null

in strument, the pressure in the gauge jacket necessary to return the 1 11 pointer to its 1 zero position was read on a closed tube mercury manometer with a cathetometer. For the solid, which had an unexpectedly low vapor pressure, the jacket was evacuated to a known pressure which was held constant, and the relative displacement between the pointers was measured as a function of the temperaillre of the eystem. The gauge was then cali­ brated in an absolute manner with a McLeod gauge, using the sensiti­ vity previously determinedo Zero-point corrections were determined by cooling the gauge to ice temperatures and determining the pressure required to "balance

• the gauge11 These corrections were negligible at higher pressures, but at the pressures exerted by solid technet ium heptoxide they amounted to 20 percent or greater. While some uncertainty was in­ troduced thereby, a reasonable estimate of the vapor pressure of the solid could be made. The pressure readings were accurate to ¼ 0.05 mm. for the liquid and to � 0.005 mm. for the solid. Temperature measure-

(5) Smith, w. T., Jr. , Line, L., and Bell, w. A., -J. Am . Chem. Soc., in press, 1952. - 25 ments were made with N. B. S. calibrated thermometers read to f 0.1° and were corrected for immersion. The gauge and its pressure jacket were rigidly_ suspended ver­ 6 tically in an oil-thermostat containing silicone oil. This thermo­ stat could be regulated continuously from below room temperature (by cooling coils) up to about 250°. (At the higher temperatures, _E-aminophenol was added to prevent polymerization of the oil.) The average temperature control was somewhat better than f 0.2°. The pressure system used for maintaining and reading the . .. pressure in the gauge jacket contained a 20 1. carboy as a damping device to minimize any sudden changes in the pressure of the system. The vapor pressure apparatus is shown in Figu.re 5. No decomposition of technetium heptoxide was noticed during the vapor pressure measurements. This was in accord with previous experience which had indicated that the compound is stable even at its boiling point (310°)0 There was no noticeable curvature to the line obtained when the data (Tatim IV, V) were plotted as log P against 1/T. A least squares method was used, therefore, to obtain the two-term Clausius-Clapeyron equations :

Solid-vapor : log P - -7205/T f 18. 279 (III-1)

Liquid-vapor : log P = -3571/T f 8.999 (III-2) The probable error for the liquid was f 1 per cent; for the solid, :I: 8 per cent. Equilibrium value s were insured by taking measurements

(6) Type DC-550, obtained from the Dow Corning Corporation, Midland, Michigan. /ft', GAUGE

----6 0 TRAVELLING:0 MICROSCOPE

�

20- LITER RESERVOIR TO VA CUUM PUMP

OIL THERMOSTAT � CLOSED TUBE MANOMETER

FIGURE 5.- VAPOR PRESSURE APPARATUS

°'I'\) 27 TABLE IV

THE VAPOR PRESSURE OF SOLID TECHNETIUM HEPI'OXIDE

a b C Tobs. Tcorr. 1/T x 103 Pobs. Pcorr. pcalc- t::/

391. 7 391.2 2.556 0.995 o.65 0.74 -0.09 389 .2 388.7 2.573 0.700 0.46 o.55 -0.09 388.2 387.7 2.579 0.660 0.42 0.50 -0.08 384.6 384.1 2.604 0.720 0.36 0.33 0.03 e 379.0 378.5 2.642 o.4oo 0.215 0.18 0.04 378.4 377 .9 2.646 0.555 0.18 0.16 0.02 373 .6 373.2 2.680 0.415 0.095 0.094 o.oo 368.5 368.5 2.714 o.24oe 0.05 0.05 o.oo 363.0 362.2 2.761 0.300 0.03 0.03 o.oo

aDegrees absolute.

�illimeter s of mercury. 0 calculated from equation III-1. d = 6 P corr. - Pcalc. ePressures so indicated obtained on cooling. TABLE V 28

THE VAPOR PRESSURE OF LIQUID TECHNETIUM HEPTOXIDE

a c 3 b d Tobs. T corr. 1/T x 10 Pobs. Pcorr. Pcalc. 6

e 393.3 393.3 2.543 0.50 1.0 o.8 0.2 . 413 .6 413 .1 2.421 1.60 2. 1 2 .3 -0.2 f 426.8 427.5 2.339 3.50 ·4.o 4.4 -0.4 440.0 441.0 2.268 6.80 7.3 7.9 -o. 6 445.3 446 .4 2.224 9.10 9.6 10 .0 -0.4 458.3 459.8 2.175 16.20 16 .7f 17 .1 -0.4 472 .8 473.8 2.111 28 .50 29.0 28.9 0.1 483.5 483.5 2.068 42.40 42 .9 f 41.1 1.8 489.7 489.7 2.042 54 .10 54.6 50.9 3.7 503.0 503.0 L988 82 .40 82 .9 f 79.4 3.5 512.5 512.5 1.949 107.35 107 .8 107 .4 0.4 519 .1 519.5 1.925 132.90 133.4f 133.4 o. o 521.0 521.5 1.918 141.40 141.9 f 141.2 0.7 524.0 524.5 1.907 153.60 154.1 154.7 -0.6 528.7 529.4 1.889 176. 70 177.2 179. 2 -2 .0

8negrees absolute • C>J1illimeters of mercury • cCalculated from equation III-2 . d 6 = P corr. - P calc. eTriple point • £pressu res so indicated obtained on cooling. 29 on both heati ng and cooling of the sample. The data are also plotted in Figure 6, along with those of Smith7 for rhenium heptoxide. The thermodynamic quantities that can be ca;J.culated from the vapor pressure equations above are given in Table VI, along with the corresponding values for rhenium heptoxide. As was expected from the similar�"atomic radii of technetium and rhenium8 as well as the is omorphous nature of pertechnetate an:i perrhe­ na.te ions9, the heats of fus ion and vaporization of the two heptoxides were found to be very similaro The most surprising observation is the fact that technetium heptoxide melts at such a low temperature, and remains liquid over a wide temperature range. This also is evidenced by the very large entropy of fusion, which app ears to be higher than that recorded for any other compound.lo The fact that technetium heptoxide solid is electric ally conducting at temperatures near the melting point is not in conflic t with this high entropy change. The reasons for the apparent non-conductance of the liquid state are, at present, unlmowno Apparently, then, the entropy change must either be associated with this conductivity or du e to a restricted rotation in the solid state with "unfreezing11 of the rotational entropy

(7) Smith, w. T., Jr., Line, L., and Bell, w. A., .!L!_ Am. Chem. Soc., in press (1952). (8) Mooney, Rose C. L., Acta Cryst o, �, 161 (1948) 0 (9) Zachariasen, W. H., Oak Rid ge National Laboratory Memo., CF-47-11-464 (1947) .

(10) Kelley, K. K., _!!..:,.§.:_Bur. Mines Bull., 393, 134 (1936) 0 DEGREES CENTIGRADE 150 125 100 75 350, 300 250 200 1000 3.00 +.,:--;_.__ +.�.,., ...... s l:J · '<:r',.• l:J• ,> ...... ,.,. .. ', ',, ...J6' , U/Q o o o ...... ', ...... � OPEN CIRCLES = COOLING

O ')\,,\ ' M. P 0300 100 2.00 \

_-3868 \ "-..0 LOG P( f +8.989 ) T \ "-• 3571 LOG - + 8.9 99 ,, "i>l " T \ '....._0 / '\ '--1_. d-1.00 \ o, 10.0 ' E ...... E - = 7320 + ), a..- LOG P( ) 15.010 \ E S T E "o� 0 t9 1 a..- 0 (R�2 O 7 -SMITH) -P.= 20 � 1.00 _J 0.00 •i ·t, o.

• 0. 10 -1.00 0 LOG -7205+ 18.279 P( S) T -� \ 0

\

-2.00 0.01 1.50 1.70 1.90 2.10 2.30 2.50 2 .70 2 .90 3 J_T X 10 FIGURE 6.- VAPOR-PRESSURE OF TECHNETIUM HEPTOXIDE \.,..) 0 31

TABLE VI

COMPARISON OF THERMODYNAMIC PROPERTIES OF TECHNETIUM AND RHENIUM HEPTOXIDES

a b Tc207 Re207

- - T 6,F;ubl. 32,950 70.4 T 3,J,500 55.6 6Fvap . 16,JJO - 28.0 T 17,700 - 28.0 T 6Ffus. 16,620 - 42.4 T 15,810 - 27�6 T 6 Hd 16 .6 15.8 f 0.1 fus. ± o.5 f f 6 ffvap. 16.3 0.2 17.7 0.1 8 6s f fus. 42 i 2 27.6 0.2 6 5vap. 28.0 ± 0.3 28.0 ± 0.1 M.P. 120 ± 1° JOO :i 0 1 B.P.f JlO f 1° 360 ± 1°

aCalculated from equations III-1 and III-2 . brrom the data of Smith, w. T., Jr., Line, L., arrl Bell, w. A. , --J. Am. Chem. -Soc., in press (1952). 0 calories mole-1. dKcal. mole- l. 8 Calories mole-1 degree-1• fcalculated from vapor pressu re equation. 32 on fusion. The complete understanding of this phenomenon must await absolute entropy determinations 3 more accurate conductivity, density and crystallographic measurements on the heptoxide o

B. The Vapor Pre ssure of Pertechnic Acid

The vapor pressure of pertechnic acid was run, in a manner similar to that describe d ·ab ove, to determine its stability. The com­ pound was prepared in situ by distilling technetium heptoxide into the lower part of the gauge and then letting water vapor into the vacuum line. The ox ide immediately becrure moist and the excess water was pumped out by warming the compound slightly. Samples prepared in this manner contained some excess technetium heptoxide , but the amounts were so adjusted that enough acid was left to give equili ­ brium pressures of wate r vapor o

The dissociation equilibrium can be wri tten :

(III-3)

The vapor pressure data for this disso ciation are given in Table VII, and are shown in Figure 7 along with similar dat a for perrhenic acid by Smith . 11 They were fitte d to a two term Clausius-Clapeyron equation for the reaction given by equation (III-J)g

log P:-2409 /T f 80201 (III-4)

9 w. (11) Smith ., W. T., Jr. , Line� L. , and Bell A. , ---J. Ain.. ·Chem. �, in press, (1952). 33

TABLE VII

THE DISSOCIATION PRESSURE OF PERTECHNIC ACID

C a 3 b d Tobs. Tcorr. 1/T x 10 pobs. Pcorr. Peale. ,6.

29lo5 291.8 3.427 3.ooe 1.00 1.00 o.oo 305.5 305.7 3.271 4.4oe 2.40 2.37 0.03 9 313 . 7 313.8 3.187 5.80 3.80 3.77 0.03 322.3 322.4 3.102 7.95 5.95 6.01 -0.06 33005 330.6 3.025 11.1oe 9.10 9.20 -0.10 337.0 337.2 2.966 15 .10 13.20 12.74 o.56 341 .4 341.5 2.928 17 .45 15 .45 15 .70 -0.25 347 ,4 347 .5 2.878 22.8oe 20 .80 20.69 0.11 353.6 353.8 2.827 29.30 27.30 27.15 0.15 9 363.3 363.3 2.753 43 .45 41.45 41 .22 0.23 aDegrees absolute.

�llimeters of mercury. 0 calculated from equation III-4. d 6 = pcorr. - Peale. e Pressures so indicated obtained on cooling . DEGREES CENTIGRADE 100 75 50 25

• HTc04(SATJ • HTc0 2.00 4 (SJ 0.100 :j::::::...... OPEN SYMBOLS = COOLING ...... --� ...... �, .

E E �A'�Q _ ',, ·�- -2374.8 E ' � -- LOG Cl..� 1.00 ,,, � P(m n)- T <.9 , + 8.2006 0.010 ...S ' ,, � [ -2394. a.. ,' , LO = 6 +8.20 ,, -, •---..... G P(mn) T 74 � ',,_---.... '-... ',, ---...... ---...... �·--.... -, ---...... ' ---...... •---...._,,,+ ---...... '--- ...... --- .... , 0.00 ---...... ---', ...... 0.001 ...... ---.... -2409 LO G + 7.680 /',, P( mmt T (HRe04 (SJ - 1 _SMITH ) .00 2.60 2.80 3.00 3.20 3.40 3.60 3 _!_ X 10 T w FIGURE 7.- VAPOR PRESSURE OF H 0 OVER HTc0 (S) AND HTc 0 ( ATl .i::-- 2 4 4 S 35

The vapor pressure of Tc207(s) itself was negligible at these tempera­ tures. It can be seen that the acid is more unstable with respect to its loss of water than is perrhenic acid . An evaporation of an acidic solution of Tc(VII) at water bath temperatures would soon give the anhydrous oxide. Continued heating at 100° would eventually give rise to loss of the heptoxide which has a vapor pressure of about o.6 nun. at this temperature. Hence, this type of procedure for con­ centration of technetium solutions is to be avoided except in basic media . By subtracting the heat of vaporization for water12 from equation (III-3) above, we can write: (III-5) The thermodynamic quantities which can be calculated from the data for equations (III-3) and (III-5) are given in Table IX.

C. The Vapor Pressure over Saturated Pertechnic Acid Solutions

The vapor pressure of water in equilibrium with saturated per­ technic acid solutions was also determined in the same manner as for anhydrous pertechnic acid. Technetium heptoxide was allowed to absorb enough water to give rise, visibly, to solution in contact with the solid pertechnic acid. The vaporization equation can be written :

(12) National Bureau of Standards ., "Selected Values of Chemical Therroc>dynamic Properties", Series I, 1948. ( II I-6)

The data were fitted as befo re to the equation:

log P : -2375/T + 8.201 (III-7)

By subtra ct ing the vaporization for water, the thermodynamic functions for the following reaction can be calculated :

(III-8)

The thennodynamic data for this system are give n in Table IX , and the vapor pressure data in Table VIII and Figure 7. The vapor pressures of the saturated solution are only a few mm. higher than for the anhydrous acid. Since the free energy difference between the solid acid and the saturated solution is given by:

(III-9) it is seen that there is verff little tendency for the system technetium heptoxide-water to add additional water after once forming the anhy­ drous perte chnic acid . Corresp ondi ng measurements for saturated perrheni c acid have not been made ; presum3.bly the situation here would be very similar. 37

TAB:i.E VIII

THE VAPOR PRESSURE OF WATER OVER SATUR.f,.T ED PER'.f.'ECHNIG ACID SOLUTIONS

b i 6::l Ta Tcorr. 1/T x 103 Pobs. Pcorr. P8alc.

312.6 312 .9 3.196 6.40 4.1�0 h.08 0.32 324.3 324.5 3.082 9.23 7.23 7.61 -0.38 329.2 329.1 3.039 ll o35e 9.35 9.63 -0.28 330.5 330.6 3.025 12 .65 10.65 10 .40 0.25 334.3 334.2 2.992 lJ .80 11.80 12.45 -0.65 337.1 337 .0 2 .967 16 .15 14.15 14.28 -0 .13 339.6 339 .5 2.946 1e.1oe 16.10 16 .01 0.09 344. 8 344.7 2.901 22.05 20.05 20 .48 -0.43 344. 8 344. 6 2.902 22.30 20.30 20.36 -0 .06 347 °7 347 .5 2.878 25.25 23 .25 23.22 O.OJ 351.1 351.0 2.849 29 .6oe 27.60 27.22 0.38 355.9 355.8 2.811 35 .50 33.50 33.49 0.01 357 .1 357.2 2.800 38.1oe 36 .10 35-58 0.52 e 361.1 361.0 2.770 4L.35 !12 .35 41.92 o.43 363.5 363 .6 2. 750 49ol5e 47.15 46 .78 0.37

aDegrees absolute . btti.llimeters of mer cury . ccalculated from equflt:i.on III-7 . d 6 = Pcorr . - P calc. bfressures so indicated obtained on cooling . 38 TABLE IX

THERMODYNAMIC DATA FOR SOLVATION REACTIONS OF TECHNETIUM .MID RHENIUM HEPTCXIDES

a 0 b Reaction 6J½50 6F250 6S250

C 2HTc04(s) = Tc207(s) + H20(v) 10 ,950 f 50 3686 f 18 24. 4 :i 0. 2 d 2HTc04(s) = Tc207(s) + H20(1) 430 ± 50 1632 f 18 -4.0 f 0.2

e HTc04(sat) = HTc04(s) + H20(v) 10 ,860 f 100 36o5 f 18 24.3 f 0 . 4 Hl'c04(sat) = HTc04(s) + H20(l) 340 ! 100 1551 f 18 -4.1 f 0.4

2HRe04(s) = Re207(s) 11, 020 ± 100 4460 ± 18 22.0 f 0.4 2HRe04(s) = Re207(s) 500 ± 100 2418 f 18 -6.4 f 0.4

acalories. bcalories degree-1• ccalculated from equation III-4. dnata for vaporization of water from Kelley, K. K., United States Department of the Interior, Bureau of Mines Bulletin 383, 1935, p. 51. ecalculated from equation III-7 . fcalculated from data of Smith, w. T. , Jr., Line, L. and Bell, W. A., cL!._Am. Chem. Soc., in press (1952). CHAPTER IV

CALORIMETRY

A. The Heat of Combustion of Technetium Me tal

One of the most fundamental thermochemical measurements which can be nade on technetium is its heat of combustion in oxygen to give technetium hepto:xide. The heats of fo rmation of technetium heptoxide an::l. aqueous pertechnic acid can be calculated from such nBasurements together with heats of solution and dilution. Similar measurements have been made on rhenium by Roth and Becker. 1 These authors found it necessary to burn the metal wit h a paraffin oil accelerator, as the combustion process would not proceed rapidly to completion otherwise. The same procedure has been used successfully with technetium. The calorimeter (see Figures 8 an::l. 9) consisted of a one liter glass dewar fitted to receive a combustion bomb and contained water as the calorimeter medium. The temperature rise was measured with a calibrated , glass-enclosed platinum resistance thermometer.

A top cover plate containing a rubber 0-ring was bolted onto a flanged ring cemented to the dewar , thus providing a water-tight seal.

Three vertical copper tubes pierced the co ver plate to allow entr ance of a resistance thermometer , glass stirring rod and ignition wire

(1) Roth, W. A., and Becker , G., � physik. _91e m ., Al59 , 27 (1932). 0 [O 0 0

Q

-9 Q _Q

Cb '-

FIGURE 8.- COMBUSTION CALORIMETER

cm g 0

0 0

FIGURE 9.- COM BUSTION CALORIMETER

cm 42 into the calorimeter. With such an arrangement, the whole assembly could be submerged in a constant temperature bath.

The combustion bomb (see Figur'e 10) was one commer cia lly available from the Parr Instrwnent Company. It was of the semi-m icro type, containing an inlet-outlet valve an:i two ignition terminals . It was constructed out of a , , steel alloy, and the internal volume was 50 ml . The bomb was susperrled in the ca lorimeter by strings attache d to the cover plate.

The sample to be burne d was pla ced in a small quartz cup held in a platinum wire ring suspended from the bomb he ad. The dewar was filled by weighing in the pr oper quantity of water . To insure a con­ stant heat-leak from the calorimeter, a radiation shield was also pr ovided. This shield consisted of an inverted co pper cup which fitted snugly into the dewar confining the heat leak between two copper sur­ faces about 2 cm. apart. This design is in accor d with the recommen­ da tions of White.2 The shield contained holes for the thermometer, st:irrer, ignition wire and strings. Hanging vertically from tre shield were two co pper fins which served to provide two sides of a stirring channel, with the bomb and dewar forming the other two sides.

A brass pr opeller on the end of a glass rod was centered in this channel and provided for the rapid mixing of the water. The propeller was ro­ tated at 1500 r. p. m. by a synchronous mot or . In order to cover the sensitive volume of the platinum resistance thermometer, it was

11 (2) White, w. P. , 11The Modern Ca.lorimeter , Chem . Catalog Co ., New York, N. Y., 1928, p. 169. 43

Fl RING ELECTRODE

------FUSE ELECTRODE

A�==-=:::,---t0r- QUARTZ CUP HOLDER

FIGURE IQ.­ MICRO COMBUSTION BOMB 44 immersed about 2 cm. into the calorimeter medium. The calorimeter assemb]y was totally submerged in a water ther­ mostat bath at 25° which was regulated to ! 0.002°. The heat leak of the calorimeter to the bath was about 25 micro-deg. seco-1 deg o -1. _ The stirrer also introduced a constant heat input of about 25 micro- d eg. sec.-1 • The water accounted for 73 .3 per cent of the total heat capacity of the calorimeter. The temperature rise was determined by measuring the change in resistance of the platinum resistance thermometer with a Mueller bridge.

Resistances were read to f 0 0 00005 ohms, corresponding to about ± 0.0005° , The temperature rise of the calorimeter was determined by the Dickinson method3 whereby the temperature difference between the two extrapolated flat portions of the temperature-time curve is taken at that time which corresponds to o.6 of the total temperature rise. (See Appendix III, Figure 15). The average temperature of the com­ bustion measurene nts was taken as the absolute temperature at the o.6 time. Since this average temperature was within a half degree of 25° in all cases, all of the heats are referred to 25°. Uncer­ tainties caused by this assumption are negligible. Quantities ranging up to 100 mg. of technetium metal prepared by ammonium pertechnetate reduction were burned together with 50 mg. of paraffin oilo The combustion oxygen pressure was 30 atm. The mixture was ignited usi ng a platinum wire over which a measured length

(3) Dickinson, Ho C., Natl. Bur. Standards Bull., 11, 189 (1914). 45

of a cotton thread fuse was hung, resting on the oil-metal combustion charge. To ignite the mi xture, a pulse of current was sent through the wire, and, although it was not sufficient in itself to melt the wire, the additional heat from the fuse did so in a fraction of a second (ca. 0.2 sec.). The heat input of the ignition proc ess was determined by separate expe riments . The nitric acid, always formed from traces of in the o:xygen used,amounted to 0.0131 m.e., corresponding to a temperatu re ° rise of 0.00016 .4 Since this same amount of acid was always forred regardless of the size of th e oil cha rge, it was assumed tha t this corresponded to the amount of acid also fornBd in the oil-technetium ° burning . The fuse and nit ric acid correction was 0.0047 ± 0.0001 . The heat of combustion of th e oil itself was determined by many separate experiments (See Table X ) • The standard deviation re- ported for the values in this chapter are calculated from the ( 2 eqllation S =l / ?f 6.x) where Z (,6.x)2 represents the sum of V n(n - 1) the sqllares of the deviation of each determination from the average of n determinations. This error function has been recommended for calorimetry> and is in common use todey. For a precision of the order of a few-hundredths cf one percent, it would have been necessa ry

(4) Washbu rn, E. w., � Research Nat. Bur. Standards, 10, 525 (1933).

(5) Rossini , F. D., and Deming, w. E., J. Wash. Acad. �, 29, 416 (1939). 46 TABLE X

HEAT OF COMBUSTION OF PARAFFJN OIL ACCELERATOR

Temperature Rise (a) 6R 6T /::.Tc orr o Sample Wt. per g.

° ° 0005840.0.. 005420 005373 68066 mg o 7.826 deg. g. -1 0005810 0.5400 Oo5353 68032 7.835 0008829 008183 0.8136 103.00 7.996 0012234 1.1381 1.1334 142042 7.958 0.05163 0.4798 0 ..4751 59.89 7.933 0.06820 0.6344 006297 79.54 7.917 0.08334 0.7745 0 ..769 8 97.65 7.884 0007950 0.7395 0.7348 92.77 7.921 0008941 0.8302 o.8255 104.20 7.922

Average(b) : 70910 .i 00018 dego g.-1 aCorrected for fuse ignition and nitric acid formation brhis corresponds to a heat of combustion (not reduced to standard state conditions) of 11.04 kcal o g-1 • 47

to correct each of the oil combustions by the Washburn corrections6 to standard state conditions. However, since the maximum precision of this calorimeter (because of its necessarily small design) was only about 0. 2 percent, these corrections were not significant.

The heat capacity of the calorimeter was determined by com­ bustion of samples of benzoic acid provided by the Bureau of Stan- dards for this purpose (Sample 39f). (See Table XI.) In this case, the Washburn7 corrections, although small, were made. Thus, the standard heat of combustion of the acid sample was quoted to be 1 2604284 international kilojoules g.- ; corrected this becane 26.4279 8 int. k j. g.-1, or 6,316.42 defined cal. g.-1• Using 4.521 deg. g.-1 from Table XI, we calculate the heat capacity of the calori­ ° meter to be 1397.1 cal. deg. -l at 25 . In the case of the technetium-oil mixtures, two corrections were made. First, the temperature rise caused by the ignition pro­ cess was subtracted (0.0047°), arrl second, the rise due to the a..� ount of oil used was subtracted. As in the case of the combustion of pure oil alone, the Washburn corrections should also be applied for both the technetium and oil. However, since the oil and technetium were burned under the same conditions as the oil alone, these coITections

(6) Washburn, E. W. , loc. cit. (7) Washburn, E. W. , loc. cit.

(8) See Appendix, Part II, for a list of the physical con­ stants used in calculations in this dissertation . 48 TABIE XI

BENZOIC ACID COMBUSTION CALIBRATION

6.R Temperature Rise 6.Tcorr. Sample wt. per g. (a) 6.T . -1 o.06330J1. 0.5891° o.5844° 128.86 mgo 4.535 deg. g. 0.03632 0.3370 0.3324 73 . 72 4.509 0.05646 0.5237 0.5190 114.68 4.526 0.05914 Oo.5489 o.5442 120.53 4.515

Average: 4.521 :I: 0.006 deg. g.-1

acorrected for fuse ignition and nitric acid formation. 49 were negligible. The combustion data for technetium-oil mixtures and the resulting corrections to obtain the heat of combustion at constant volume of technetium are given in Table XII. It should be noted that these data are for conditions existing in the bomb, and further, that corrections must be made to obtain the true standard state enthalpy of combustion. The technetium heptoxide from the technetium combustion would have dissolved in the water formed from the combustion of the oil to give a concentrated solution of pertechnic acid. Because of the limited quantities of technetium available, heat of dilution studies could not be made on these relatively concentrated solutions . There­ fore, 2.00 additional ml. of water was placed in the bomb so that the final product of the combustion was a more dilute solution of per- technic acid. This solution was then used for heat of dilution studies which are reported in a later section of this chapter. From Table XII, we obtain for the bomb he at 0.998 deg. g.-1. Using a heat capacity for the calorimeter (from page 47 ) of 1397.1 -1 9 cal. deg. , and an atomic weight for technetium of 98.91 , we ob- tain for the actual bomb process:

(IV-1) 1 = 2HTc04(0.3 M.) 6E -276.o ± 2.8 kcal.

(9) See Appendix II for the value used for the atomic weight of technetium in this research. TABLE XII

HEAT OF COMBUSTION OF TECHNETIUM METAL

- - (a) Temperature Rise ( ) per g. 6.R 6.T 6T(corr.) Wt. Oil 6 T (Oil) 6.T(net) wt. Tc b . 0.07150Jl o.6645° o.6598° 79.56 mg . 0.6293° 0.0305° 31.01 mg. 0.983 deg. g. -1 0.08422 0.7834 o. 7787 94.83 0.7501 0.0286 29.25 0.979 0.09801 0.9117 0.9070 107.44 o.8499 0.0572 56 .39 1.014 0.12368 1.1526 1.1479 135. 26 1.0699 0.0780 76.25 1.023 0.06718 o.6244 0.6197 72.78 0.5757 0.0440 4.5 .38(c) 0.969 (c) 0.13433 1.2492 1.244.5 144.60 1.1438 0.1007 99.oo 1.018

Average: 0.998 :! 0.009 deg. g '"'-1

aCorrected for fuse ignition and nitric aci� -format-ion. bweight of technetiwn metal actually burned; this represents 80 - 90 percent of the initial charge. c ° 3 samples so indicated had been degassed at 800 far 2 - hours in vacuum. � 51

The concentration of 0.3 �- represents the average concentration for the six experiments given in Table XII . Since there was no notice­ able trend in the heat of reacticnas a function of the amount of metal burned, it was concluded that the differences in concentration of pertechnic acid solutions forned did not contribute any significant error. It was found that 80-90 percent of the technetium metal in the charge was burned, and that the residue left in the ignition cup was pure technetium metal . This observation was checked by determining the actual amount of technetium present as pertechnic acid in the bomb solution. The average distribution of technetium was as follows : on the bomb head, 10 percent ; on the bomb walls, 5 percent ; in the bomb solution, 85 percent . Since most of the technetium was found in the bomb solution, the average concentration of this solution will be taken as the average concentration of the pertechnic acid formed in equation (IV-1) . Two of the sample s of technetium metal were thoroughly degassed at high temperatures in a vacuum following their preparation from ammonium pertechnetate by hydrogen reduction. Since the heats of com­ bustion of these two samples were, within the experimental error, the same as for the other, non-degassed samples, it may be concltrled that technetium prepared in this manner cbes not contain hydrogen to any appreciable extent. To correct equation IV-1 to standard state conditions, a heat term for the change of pressure from 30 atm. to unit fugacity must be added: 52

0 (IV-2) 7/2 02(g , unit fugacity) = 7/2 2(g , JO atm.)

Taking (aE/aP)250 equal to -6.5 joules atm.-1 mole-1 for the com­ 10 pression , we obtain 0 068 kilo-joules or 0.16 kcal. for equation IV-2. Hence�

aq. 2 Tc(s) + H20(1) + 7/2 02(g J unit fugacity) f

= 2HTc04( 0.J .!1•) (IV-J) 6E = -276.2 f 2.8 kcal.

(The heat of dilution of O.J �· pertechnic acid solution to infinite dilution will be described in a later section of th is chapter.) Since the actual energy measured for the bomb process corres­ ponds to a change in E, we :nm.st add to this the PV work done to obtain the enthalpy change:

6H = 6E f 6nRT (IV-4)

Taking .6n as -7/2, 6.nRT has the value of -2.1 kcal. The heat of formation for the pertechnic acid solution then becomes :

2Tc(s) + H20 (1) + 7/2 02(g) = 2HTc04( 0.J �-) (IV-5) 6 H = -278.J :f: 2.8 kcal.

The heats of combustion of rhenium metal and rhenium trioxide have also been determined. The data for rhenium can then be compared

(10) Rossini, F. D. , and Frandsen, M., !!.!. Research � Bur. Standards, 2._, 733 (1932). 53 with the heat of combustion of the met al and Re03 determined by Roth 11 and Becker as a check on the accuracy of the experimental method.

These data and the comparison are given in Appendix IV.

The agreement is satisfactory for rhenium metal, but not in agreement wi th their value for Re03 . Si nce this later value was determined only indirectly, it is felt that the value determined in this re­ search by direct oxid ation is more accurate.

B. The Heat of Solution of Technetium He ptoxide

Since it was necessary to use oil to produce a sat is factory combustion of technetium metal samples, the fina l state of the technetium was that of an aqueous solution of the oxide, i.e. , a solution of pertechnic acid. Before the heat of formation of the oxide can be determined, therefore, it is necessary to know the heat of solution of the oxide to give the same concentration of acid as that resulting from the combustion (O.J M.). An alt ernative pro­ cedure would be to determine the heat of solution of the anhydrous oxide at infinite dilution, and the heat of dilution of the bomb solution to infi nite di lution.

The solution calorimeter was oonstructed from a small cylin­ drical dewar (see Figure 11 ) 5¼ inches long and l½ inches in diameter . It was fi tted wi th a standard taper joint so that a glass stoppered top could be fitted onto it. Two glass tubes were sealed

(11) Roth, w. A., and Be cker, G., z. physik. Ch�, A�59 , 27 (1932). 54

I V

. V -

� " � � I� I �' ' � � ' §

II 11: II II<>

I� c�I � �

FIGURE II.- SOLUTION CALORIMETER 55

vertically into the top which provided inlets for the platinwn resistance thermometer and the glass stirrer used to stir the con­

tents of the calorimetero The calorimeter was submerged in a thermo­ stat bath controlled to ± 0.002° 0

About 100 - 200 mgo �f technetium heptoxide was sealed in a small thin-walled glass bulb which was cemented to the end of the platinum resistance thermometer. The bulb could then be broken by applying a slight pressure on the top of the thermometer. The stirrer was rotated at 250 r. p.mo by a synchronous motor. The calorimeter contained J2.0 ml. of water which served as both solvent and as a calorimeter medium. The re·sistance of the platinwn resistance thermometer was measured as before, except that the in- , ° dividual measurements were determined to ± 0.0001 ohms, or ± 0.001 . The temperature rise was about 0.3°, so that this still gave a precision of about 0.3 percent. The total temperature rise was measured by taking the rise at the 0.6 time as in the combustion calorimetry. The calorimeter was calibrated by the neutralization of an aqueous sodium hydroxide solution with a slight excess of hydrochloric acid to form water. The sodium hydroxide solution was weighed into a small glass bulb arrl broken in the hydrochloric acid solution. The heat data for sodium hydroxide, sodium chloride and hydrochloric acid 12 were taken from Harned am Owen. Corrections were made for the

(12) Harned, Ho s., ar:rl Owen, B. B., 11The Physical Chemistry of Electrolytic Solutions", Reinhold Publishing Co., New York, N. Y., 1943, p. 541. 56 heats of dilution. During the measurements Calorimeter I was acci­ dentally broken and a secon::l one had to be constructed. The agree­ ment (see Table XIV) between values obtained from the two calori­ meters was therefore gratifying. From the calibration data of TablelllI, and the heat data from Table XIV, the heat of solution at 25° of the oxide was estimated:

H20(1) + Tc207 (s) + aq. = 2Hrc04(0.04 .&) (IV-6) 6 H = -11.59 i 0.08 kcal.

Since this solution is so dilute, we shall assume that this heat also represents the heat of solution of the oxide to give an infinitely dilute solution.

c. The Heat of Dilution of a Pertechnic Acid Solution

Since the solution of pertechnic acid formed in the combustion bomb experiments was still relatively concentrated, it was necessary to determine the heat of dilution of this solution to essentially in­ finite dilution, to see if an appreciable amount of heat were involved. The heat of solution calorimeter previously described was used. Two ml. of a 0.J28 M. pertechnic acid solution was sealed in a small glass bulb and broken in the calorimeter containing J0.00 ml. of water to -o.oo,5° give a solution which was 0.020.5 � The heat change of corre­ sponded to the absorption of 0.28 kcal. mole-1 of acid . As before, it was assumed that this latter concentration of pertechnic acid is essentially infinitely dilute. TABLE XIII

CALIBRATION OF SOLUTION CALORIMETER

-(d) Calorimeter Loading(a) M.e . H 0 Formed (b) c(c) 2 6,H 6T corr.

I ° 32.00 0.994 -12,981 0.2037 39.77 39.17 I 31097 0.971 -12,981 0.3184 39-57 39-6o

Average: 39 .69 :i .09

II 31.86 0.862 -13,000 0.2833 38.92 38.82

aGrams of H20. bcorrected for heats of dilution of NaOH, HCl, NaCl. Cin cal. deg.-1 at 25°.

° d Vl Adjusted to loading of 32.00 ml. at 25 . -.J 58

TABLE XIV

HEAT OF SOLUTION OF TECHNETIUM HEPTOXIDE

Calorimeter b..T Cone.Ca) -L':.H soln.(b )

I 215.6 mgo 0.2045° 0.0448 11.67 II 182.0 0.1741 0.0368 11.51

Average: 0.0408 11.59 ± 0.08

aConcentration of resulting solution in moles 1.-1 • b in kcal. mole-1 of Tc207. 59

The heat of combustion of technetium to give anhydrous tech­ and netiwn heptoxide pertechnic acid at infinite dilution (HTc04(aq)) can now be calculated . From equation (IV-5):

t 2Tc(s) + 7/2 02(g) H20(l) f aq. = 2HTc04(0.3 M.) ( IV-5) 6H = -278.J ¼ 2.8 kcal.

Adding to this

t 2HTc04(0.3 !i:_) aq. = 2HTc04(aq.) (IV-7) 6H : t 0.6 kcal .

2Tc(s) 4 7/2 02(g) t H20(1) + aq. = 2HTc04(aq.) (IV-8) .,6 H = -277. 7 ± 2. 8 kcal.

is obtained. Adding

+ + aq. 2HTc04(aq.) = Tc207(s) H20(l) (IV-6) 6,H = 11.6 i 0.1 Kcal.

to (IV-8) gives :

2Tc(s) + 7/2 02(g ) = Tc207(s) (IV-9) 6H = -266.1 ± 2.6 kcal.

Taking the value of 68,317.1 cal.mole-1 for the heat of formation of · water13 under standard conditions at 25° and adding it to equation

(13) "Selected Values of Chemj_cal Thermodynamic Properties", National Bureau of Standards, Washington, D. c., 1948, Series Io (IV-8) , the heat of formation of aqueous pertechnic acid £ran its el�nts is :

(s) f 2 + aq. ) Tc 02(g.) f ½ H2(g ) = Hl'c04(aq. (IV-10) '6.H :-173.0 ± l.7 kcal. CHAPI'ER V

OXIDATION REDUCTION CELLS

Both manganese and rhenium form dioxides which can be used to study the oxi dation potential of the elezoont from the +4 to the f7 oxi dation state. These compounds form reversible electrodes; the manganese di oxide-permanganate electrode has been discussed by

1 2 Willdey, and Hugu s has recently measured the rhenium dioxide-per­ rhenate potential.

The oxidation potential for the technetium di oxide-pertechne­ tate electrode has been measured in both acid and basic solutions for the thermodynamic characterization of technetium di oxide. Both typ es of cells were used to prove that the electrodes are reversible and that the electrode equations are as follows :

(acid) (V-1)

(base ) (V-2)

Wi th the measured potential of the basic cell, the vo ltage of the acid cell can be calculated and compared wit h the measured value. Good agreement was obtained for th ese cells.

(1) Wadsley, A. D., and Wa lkley, H., J. Electrochem . Soc ., 95, 11 (1949). (2) Hugus, z., and La timer, w. M., To appear in the ne w edition of "Oxidation Potentials", 1952. 62

A. Acid Cell

The half-cell consisted of a test tube about 3 in. long and

1 in. in diameter, fitted with a rubber stopper. Inserted through the stopper was a platinum rod with a small cup on its end to con­ tain the technetium dioxide . A glass stirrer extending into the cell

was rotated slowly during the measurements to avoid polarization of the electrode. 'l'he cell solution of pertechnic acid wa s made by

dilution of a mere concentrated, standardized solution. This concen­ tration was also rechecked after each determinati on by converting the solution to ammonium pertechnetate and weighing. The half-cell was connected by a salt -bridge of 1.8 !!.:_ potassium nitrate and 1.8 M. potassium chloride solution as recommended by Grove-Rasmussen3 to a -silver chloride half-cell of known voltage . The two half- o cells we re immersed in a thermostat at 25 . The voltage of the cell was measured on a type K-2 Leeds and Northrup potentiometer, wh ich was standardized before each reading against a stan:iard Eppley cell, cali­ brated by the National Bureau of St andards .

The salt-bridge was co nstructed of pyrex tubing in the fo rm of an inverted U-tube, the two ends of which contained sealed-in fibers, obtained from two commercial calomel electrodes . These fibers provided solution contact between the two half-cells, but prevented diffusion of any but trace amounts of bridge solution into either half-cell com ­ partment.

(3) Grove-Rasmussen, K. V., Acta. Chem. Scand. , 2, L22 (1951) . 63

The technetium dioxide used in one series of determinations was prepared by Nelson4 by the reduction of aqueous ammonium pertechnetate solution with zinc ani hydrochloric acid . For the second series of experiments, an electrode of platinum gauze was used onto which black deposits of technetium dioxide had been electroplated from ammonium pertechnetate solutions. These deposits were not characterized by chemical means, al though rhenium dioxide prepared in this manner is well lmown.5 Since the voltage of this electrode was the same within experimental error as the voltage of the other technetium dioxide elec­ trode, it was concluded that such a procedure does result in technetium dioxide. The silver-silver chloride half -cell was made from a silver wire which had been cleaned in nitric acid and thoroughly washed. It was used as the anode in an electrolytic cell containing dilute hydrochloric acid. Brief electrolysis in this cell formed a thin adherent deposit of silver chloride on the silver wire. After washing, this electrode was immersed in a hydrochloric acid solution of known concentration. The cell can be represented as :

Ag/Ag Cl : HCl : : ��3 : : HT c04 T c02/Pt (V-3)

(4) Nelson, c. M. , "The Magnetochemistry of Technetium and Rhenium� Doctoral Dissertation, The University of Tennessee, 1952; the author is indebted to Dr. Nelson for the use of this compound. (5) Druce, J. G. F., "Rhenium", The University Press, Cam­ bridge, England, 1948, p. 32. 64

The data for the cell at various acid concentrations are given in Table XV . The reaction for the entire cell can be written:

3Ag _ + 3HC1 + HTc04 - 3AgCl + 2H20 + Tc02 (V-4) for which : 0.0591 E = Eo + (V-5) n wh ere n refers to the number of electrons (3) involved in the oxi­

1 dation and the a s refer to activities of HCl arrl IITc04. Si nce the HCl concentration was kep t constant (0.05 M.), the product of the activity coefficient,6 0.8304, times the concentration,0.05,can be sub­ stituted for the activity of the HCl. This quantity must then be squared, si nce the activities referred to in equation (V-5) are mean ionic activities. Substituting nu merical values:

(V-6)

In this equation the activity of pertechnic acid has been rep laced by the appropriate molalities and activity coefficients. If it is further assumed that the activity coefficient of the pertechnic acid in very 7 di lute solutions can be considered as �

1 2 = -0.5065 M. (V-7)

(6) Harned, H. s., and Owen, B. B., 11The Physical :Chemistry of Electrolytic Solutions", Reinhold Publishing Company, New York, N. Y., 1943, p. 547. (7) Harned, H. s., and Owen, B. B. , ibid ., p. 119. 65 TABLE XV

EMF OF THE TECHNETIUM DIOXIDE-PERTECHNETATE ELECTRODE IN PERTECHNIC AC.ill SOLUTIONS

a log !1:. Eb Electrode (f) Eobs . M. -log � -0.0394 1

Tc02 (from 0.276 0.00135 2.870 0.11.31 0.552 zinc re- 0.269 0.000679 3.168 0.1248 0.557 duction) 0.251 0.000338 3.471 0.1380 0.552 0.231 0.000135 3.870 0.1520 0.546 average : 0.552 f 0.003

Tc02 0.268 0.00179 2.747 0.1082 o.S4o (electro­ 0.256 0.000897 3.047 0.1201 0.549 lytic) average : 0.545 :l 0.005

Total average: Oo549 :i 0.004 volts

ain moles 1.-1 of pertechnic acid.

l Eobs. - 0.0394 log M. + 0.1634 volts, against the Ag-AgCl elec­ �trode.= the expression for the cell voltage as a function of concentration then becomes:

0 2 E:E - 0.1634 f 0.0394 log M. - 0.01997 M. (V-8) -Hrc04 or, on rearranging,

1 E : E + 0.1634 - 0.0394 log M : E 0 - 0.01997 M.2 (V-9) 1 .HTc04

When the left side of equation (V-9) is plotted against the square root of the concentration, a straight line of almost zero slope, and an 0 intercept at zero concentration equal to E for cell reaction (V-4) , is ob tained. The cell reaction indicates that the technetium dioxide elect­ rode should be the positive electrode, and this was also verified. Calculations using equation (V-9) are given in Table XV. At concentrations above .00135 M. some deviation from the limiting behavior was noticed; hence, only the lower concentrations are considered. Part of this deviation may be due to the unknown effect of the junction potentials on the cell . voltages . 0 The average value of th e defined E1 above will be equal to E since the last term in equation (V-9) is negligible at the concen­ trations given in Table XV. From Table XV ,E1 is equal to 0.549 f 0 0.004 volts. E for the total cell reaction is the sum:

Eo o Eo E f = .rc�-Tc04 Ag-AgCl (V-10) 67

0 Using the value for E of the Ag-AgCl electrode as -0.222 volts8, the voltage for the half-cell reaction W-� beco!D9s -0.771 volts. These values are, of course , based on the usual assumptions th at the voltage of the hyd rogen electrode at unit activity is zero, and that the junction potential is negligible at these low concentrations .

B. Basic Cell

The basic cell was constructed in the same manner as the acid cell, except that the dioxide-pertechnetate half-cell contained sodium pertechnetate and sodium hydroxide of equal concentrations . These were formed in si tu by pipetting the required amounts of standardized pertechnic acid and sodi um hYdroxide solutions into a known amount of distilled water already in the cell compartment . The referenc e electrode in this case was th e me rcury -mercuric oxide elec­ trode in a sodium hydroxide solution. The cell was first constructed with the two electrod es immersed in the same solution, but this method was abandoned because of th e tendency for the fimly-divi ded technetium oxide to become mi xed in with the me rcuric oxide of the other electrode.

Therefore , the half cel ls were separated, using the same salt bridge as for the acid cell.

The technetium dioxide electrode used first was of the same ty pe as used in the acid cell . However, difficulty was experienced in getting the technetium dioxide to stay in the cup, sinc e it apparently disinte-

(8) Harned , H. S., and Eh lers, R. W., --J. Am. Chem. Soc., 54, 1350 (1932). 68

grates in basic solution int o very fine particles. Hence, the voltage

of the cell was not steady, and this electrode was unsatisfactory.

The electrode which finally gave satisfactory results was obtained by

plating technet ium di oxide onto a platinum gauze as in the case of the

acid cell. The technetium dioxide so formed is very adherent , and

serves as a satisfactory electrode from a physical standpoint . The two half-cells were immersed in a thermostat at 25°, and the voltages read as before . Some difficulty was experienced in that

the voltage of the cell drifted with time. This was possibly due to dioxide absorption by the basic solution with a subsequent

change in the hydroxyl ion concentration. However, when nitrogen

gas was passed through the cell to avoid this di fficulty, the voltage

still drifted (fell) . It had als o been observed previously in the

acid cell that when nit rogen gas was rubbled through the solution,

the normally stable voltages fell . This could be due to the nitro­ gen undergoing some catalytic reaction at the technetium dioxide

electrode, or to an electrode poison in the gas o Since the drift

was not very rapid, the dat a recorded in Table XVI for this cell were taken during the first few hours .

Below a concentration of about Ooo6 !i!_ Tc04, the cell voltage was neither st eady nor reproducible . The same effect has been ob­ - 9 served fo r the basic Reo2-Reo4 electrode o Since the technetium was not available for more concentrated solutions, it was not possible

(9) Hugus , z., and Lat imer, W. M., University of California Radiation Laboratory Chemistry Di vision Quarterly Report, UCRL-ll96, 1951 . 69 to make a concentration study. However, it will be shown that this is not necessary in the case of the basic cell, because of the form of the cell reaction. Two different electrodes were used, however, to obtain the data reported in Table XVI. The cell may be represented as:

(V-11)

The cell reaction may be written :

2NaTc04 + 3Hg t H 2 0 - 3Hg0 f 2Tc02 + 2NaOH (V-12 ) and the expression for the voltage is :

2 = _ .0591 a Na0H EB E� log (V-13) 3 2 a NaTc04

Since the concentrations of sodiwn hydroxide and sodiwn pertech­ netate were the same, then :

2 ( )(.0591) log jNaOH (V-14) 3 -YNaTc04

It can be seen that in dilute solutions,where the activity coefficients of all 1:1 electrolytes are about the same , the measured voltage should correspond to the E� for the cell reaction, (V-12 ). The cell reaction requires the technetium dioxide electrode in this case to be the negative electrode (anode), which was verified. Using the average value of 0.409 f 0.004 volts (see Table XVI) for the cell reaction and subtracting 0.098 volts for the mercury- 70

TABLE XVI

EMF OF THE TECHNETIUM DIOXIDE-PERTECHNETATE ELECTRODE IN BASIC SOLUTION

a Tc02 Electrode ,(-) Concentration Voltage b

I 0.0683 � NaTc04 0.413 f 0.003 0.0683 M. NaOH

II 0.0683 M. NaTc04 0.406 :l o.oos 0.0683 M. NaOH

Average: 0.409 f 0.004 volts

aElectrolytic deposit on platinum. bAgainst the Hg/HgO-electrode. 71

mercuric oxide electrodelO the voltage ror half cell reaction (V-2) becomes 0.311 volts :

(V-2)

E� = �0.311 i 0.004 volts

The two cells now can be compared as a test of the assumptions 0 involved in calculating the E for the acid and basic cells. To convert from the basic to acid cell voltages:

(V-15}

6E = � - EA = fl.103 volts (V-16)

0 Therefore EB(calc.) = 1.103 - 0.771 = + 0.332 volts. The mean average for both of these cells will be taken as:

� = -0.782 ¾ 0.011 volts (V-17)

EB = 0.322 ! 0.011 volts (V-18)

, (10) Macinnes, D. A., "The Princ iples of Electrochemistry11 Reinhold Publishing Company, New York, N. Y., 1939, p. 254. CHAPl'ER VI

THERMOCHEMICAL CA LCULATIONS

The heats and free energies of formation and reactions involving a nwnber of compounds of technetiwn can now be calcuiated from the data given in the previous chapters taken together with the entropy values for technetium and its compounds given in Appendix V. From the free energies or formation of the oxides, the bond energy of the technetium­ oxygen bond can be estimated. These values will then be compared to the available data on other elements of Group VII in an attempt to show, once again, that the periodic arrangement of elements is useful

in predicting the properties of a new element. All of the auxiliary data, unless otherwise specified, have been obtained from two sources: absolute entropies from the compilation of Kelley1-, and heats and free energies of formation from the Bureau of Standards Tables.2 While the thermodynamic data are given to five significant figures, this is intended for calculational purposes only. The uncertainties in values are given in Table XVIII and for the most part amount to aoout 1-2 percent.

A. The Free Energies, Heats, and Entropies of Formation of Technetium (VII) Compounds

(1) Kelley, K. K., .!!.!. .§.:. Bur. Mines Bull., 477, 1950. (2) "Selected Values of Chemical Thermodynamic Properties", U. S. Bureau of Standards, Washington, D. c., 1948, Series I, II, III. 73

From the heat of formation of solid technetium heptoxide (Chap­ and ter IV) of -266 .,100 cal. mole-1., the entropy of technetium heptoxide (Append:uc V) of 39 .0 e.u., the entropy of formation, and thus, the free energy of formation of the heptoxide can be calculated to be -222,540 cal. mol e-1 :

2T_c(s) + 7/2 0 2(g.) : Tc2?7_(s) t6,. HO : -266.,100 cal. 0 (VI-1) 6 F = -222 ,540 t6,, s0 = -J.L6.l e.u.

Using the vapor pressure data (Chapter III) far the heptoxide, the thermodynamic fW1ctions for the heptoxide vapor become: = 2Tc(s) t 7 /2 0 2(g. ) Tc207(v) .6 Ho = -233,150 cal. (VI-2) ..6FO -210,580 cal. 0 = .6S -75.7 e.u.

From the entropy of formation of the vapov, and the absolute entropies of oxygen and technetium, the entropy of the heptoxide vapor at ;25° is

109 .4 e ou. This value indic ates that the gaseous technetium heptoxide molecule llUlst have a considerable number of rotational and vib rational degrees of freedom. The heat of formation of pertechnic acid can be calculated from the vapor pressure data of Chapter III and equation (VI-l). Using ,6 H as 430 cal.,D.F as 1632 cal., and 6S as -4.0 e.u., for the decomposition 74

into the heptoxide and water, the following is obtained:

2Tc(s) + 7/2 02(g) f H20(l) = 2HTc04(s) � H = -266,530 cal. (VI,-3) � F = -224,171 cal. Lis= -142 .1 e.u. from which it follows : ½ = Tc(s) + H2(g) + 2 02(g) HTc04(s) �HO = -167,420 cal.� 0 (VI-4) Li F = -140 ,430 cal. fiSO = -90.5 e.u. For solutions of pertechnic acid the following quantities were calcu­ lated from (IV-4) , vapor pressure measurements (Chapter III), heats of solution (Chapter IV), and entropies (Appendix V) :

Tc(s) t ½ H2(g) + 2 02(g) + aq. · = HTc04(0.3 M.) (VI-5) /5,. H° = -173,300 cal.

= Tc(s) f ½ H2(g) + 2 02(g) + aq. HTc04(sat.) Li °H = 167, 760 cal. - (VI 6) 0 - Li F = -141,980 cal. .tiso = -86.4 e.u. and Tc 2 0 t ½ H aq. = c0 (s) + 2(g) 2(g) i HT 4 (aq.) '"" 0 L ..-l LiH = -17 3, 000 cal. ·. ·· 0 .tiF = -lS0,550 cal. (IV-10) 0 /5,. s = -75-3 e.u.

The cell potential of the hypothetical half-cell reaction in­ volving technetium metal and pertechnetate ion can now be calculated: 75

(VI-7)

0 E : -6F/23,061(7) = -0.472 volts

The potential of the technetium-technetium dioxide couple then becomes -o.240 volts from equations (VI-7) and (V-17). The heat and free energy of formation of potassium pertechnetate can be calculated from the solubility data of Parker3 and the data already recorded in (VI-1 through VI-6). Parker gives the Ks.p. for potassium pertechnetate as O.)ili at 27° and 0.14 at 7° . From the equation:

K S•P• (l) log (VI-8) K s•P•(2) the heat of solution is calculated to be 9,550 cal . mole-1• If we further neglect the heat of dilution of a saturated solution to infinite dilution, then:

KTc04(s) t aq. = KTc04(aq.) 6.H: 9,550 cal. (VI-9) 6.F° = 490 cal. 0 30. 6. s = 2 e. u.

(3) Parker, G. w., Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-lll6, 1951 . where the free energy of solution was calculated from:

o �F : -2.303 RT log Ks.p. (VI-10)

° No correction for the temperature difference between 25 and 270 has been made. It was shown in Chapter II that pertechnic acid behaved like a strong acid. Assuming therefore, that its heat of neutralization by potassium hydroxide at infinite dilution is -13,308 calories, the heat of formation of solid potassium pertechnetate may be calcu­ lated:

t = K(s) + Tc(s) 2 02(g) KTc04(s) � HO = -237,830 cal. {),FO = -213, 710 cal. (VI-11) 0 b. s = -80.9 e. u.

The details of this calculation are given in Table XVII. The entropy of technetium heptoxide deserves special attention.

The entropies of most of the lower oxides of technetium can be estimated fairly accurately, as was do ne in Append:ix V, by comparing the contri­

1 bution of the oxygen in similar oxides and using Latimer s tables4 for the contribution of the metal ato m. However, there is so little information on the higher oxides, that this method is subject to more uncertainty. In addition, the large entropy of fusion of technetium heptoxide indicates that some of the contributions of rotation and

(4) Latimer, W. M., !!.:._ Am. Chem. Soc., 73, 1480 (1951). 77 TABLE XVII

THE HEAT OF FORMATION OF POTASSIUM PERTECHNETATE

Reaction - 6. H

q (1) 2K(s) f 2H20(1) f a . = 2KOH(aq .) � H2(g) 93,366 calo (2) 2Tc(s) + 7/2 0 2(g) + H20(1) + aq. = 2HTc04(aq.) 277,700 (3) 2KOH(aq.) f 2Hrc04(aq.) : 2KTc04(aq.) + 2H20(l) 26,616 t q a (4) 2KTc04(aq.) = 2KTc04(s) a . 19,200 (5) H2(g) + ½ 02(g) = 8 20(1) 68,317

t 4 2K(s) + 2Tc(s) 4 02(g) = 2KTc04(s) 85,199

or 6.H1 : -242, 600 cal. mole-1

aFrom solubilit,y data of Parker (Parker, G. W., Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-1116, 1951) and assuming the heat of dilution of the saturated solution is negligible. 78 vibration present in normal crystalline solids may be peculiarly lacking in solid technetium heptoxide and consequently estimates of this nature wo uld not be justified . From what data exists, how­ ever, the entropy calculated by th e same method as the other oxides is 43.9 e.u. Another met hod which can be applied is to base the entropy upon that for solid pertechnic acid and the entropy change for :

½rc207(s) f ½H 20(1) = IITc04(s) (VI-12 ) 6S = 2.0 e.u. from Table IX . With an entropy of 29.9 e.u. for HTc04(s), the ° calculated entropy of Tc207(s) is 39.0 e.u. at 25 . The latter value is adopted because the entropy calculated from the additivity of the oxides is so uncertain.

B. The Free Energy, Heat , an::l. Entropy of Formation of Technetium Dioxide

The free ener gy, enthalpy, and entropy of formation of tech­ netium dioxide can be calculated from the cell voltages given in

Chapter V:

(V-1,17) 0 E = -0.782 volts 0 6F = 54 ,101 cal . 79

The free energy of formation of Tc02 is then:

-91,270 cal . mole-1 (VI-13)

Using the value of 14.9 e.u. for the entropy of technetium dioxide, the entropy of formation becomes -40.9 e.u., and the heat of for­ ° mation at 25 is -103 ,465 caJ.. mole-1 :

Tc(s) + 02(g) = Tc02(s) 0 6H' = -103 ,460 cal. 0 (VI-14) 6 F = -91,270 cal. 0 : 6S -40.9 e.u.

c. The Free Energy, Heat, an:l Entropy of Formation of Technetium Trioxide

Although none of the thermodynamic properties of technetium trioxide has been measured directly, the heat of formation of this compowxi can, nevertheless, be estimated with fair accuracy. The method may be based on the assumption that the difference between the heats of formation of t echnetium trioxide and technetium hep­ toxide is the same as the difference between the heats of formation of rhenium trioxide and rhenium heptoxide. This difference is 3 kcal. 1 mole- Re03 (See A IV-5). The estimated heat of formation for tech- netium trioxide then becomes:

0 l 3 0 = 2..6Hf(Tc207) - . (VI-15) = -130 i 3 kcal. mole-1 80

An estimation can also be based on the fact that technetium trioxide has been observed to disproportionate in water (Chapter II ):

3Tc03(s) f H 20(l ) + aq. = 2HTc04(aq.) + Tc02(s) (II-.5,6) b,F�O

The value of the free ener gy of formation which causes the free energy change for this reaction to be slightly negative (by 2-3 · .. kilocal . ) is -109,740 cal. mole-1 of technetium trioxide . Using 1.5. 7 e.u . for the entropy of technetium trioxide, the heat of for­ mation which will enable reaction (II-.5,6) to be spontaneous is -129,000 i .5000 cal . mole-1• Since this value is wit hin the experi­ mental error given in (VI-1.5), it is adopted for the heat of forIMtion :

Tc(s ) + 3/2 0 2 (g ) = Tc03(s ) 0 � H = -129,000 cal .

0 (VI-16) �F = -109,740 cal.

0 ,� s = -64.6 e. u.

The half-cell reactions for the trioxide-pertechnetate and the tri­ oxide-dioxide couples can be calcul�ted from the free energy of for­ ma tion of technetium trioxide . These cell reactions are , of course , hypothetical since the trioxide reacts with water; however, such voltages are of interest in estimating the strength of an oxidation ­ reduction system.

For the technetium trioxide-pertechnetate electrode : 81 = + - Tc03(s) t H2o(l) + aq. Tc04(aq.) t 2n + e

A o A o o o 0 F • 0 F - 0" F - 0" F f(Tco4,aq.) f(Tc03) f(H20) (VI-17) 0 ,1 F = 14,880 cal. 0 E = -1.L,88o/2J,061 = -o.65 volts and for the technetium dioxide-trioxide couple :

0 (VI-18) 6 F = -38,220 cal. 0 E = 38,220/2 x 23,061 = 0.83 volts

Because of the instability of technetium trioxide in aqueous solutions, neither the reduction of pertechnetate ion nor the oxi­

dation of technetium dioxide to tm trioxide is feasible. The vol­

tages indicate that such oxidations and reductions mig ht well be

carried out in non-aqueous media. The free energies also indicate why technetium heptoxide is so easily decomposed by orgarri..c materials :

0 (VI-19) 6 F = -58.4 kcal. and also : = TcOJ(s) + co (g) Tc02(s) + C02(g)

0 (VI-20) 6 F = -43 .o kcal. 82

Consequently, the oxidation of Tc02 by to higher oxides is not feasible.

D. The Oxidation-Reduction Scheme for Technetium

The oxidation-reduction scheme for technetium, technetium dioxide, technetium trioxide, and pertechnetate can be constructed following the method used by Latimer.5 Such a diagram may be used to detennine the likelihood of a given reaction occurring in aqueous media. Potential diagrams are given for both unit activity hydrogen and hydroxyl ion. Voltages in basic solutions can be calculated from the acid potentials by reason of the fact that the hydrogen ion concentration is 10-14 M. in a basic solution of unit activity. The potential diagram for technetium is given in Figure 12. The value for the Tc-1 potential is estimated from the value of 0.4 volt for the Re-1 pot ential estimated by Latimer.6 A summary of the calculated and measured thermodynamic properties of technetium and its compounds is given in Table XVIII.

E. Comparison of the Properties of Sub-group VII

A comparison of the thermodynamic properties of the other menbers of sub-group VII, manganese and rhenium, indicates that in general the properties of technetium are intermediate between

(5) Latimer, W. M. , "Oxidation Potentials", Prentice-Hall, New York, N.Y., 1938. (6) Latimer, W. M., loc. cit., p. 227 . TABLE XVIII

SUMMARY OF THERMODYNAMIC PROPERTIES OF ·TECHNET IUM AND ITS COMPOUNDS (25° )

- 0 !::. 0 o - I::. S o(b) Compound - H 1\. f 5 b. Tc 0 c al. 0 c al. 0 e.u. 6.8 f 0. 2 e.u. (s) ------Tc(g) 36.5 i 0.2 43.26 f 0.01 a f Tc02(s) 103,460 i 2000 90,270 f 2000 t 40.9 0.5 14.9 :1: 0.5 Tc03(s) 129,000 i 5000 109,740 .i 5000 64. 6 ± o.6 15.7 :i: o.6 Tc207(s) 266,100 ! 2600 222,540 f" 2600 146.1 ± 2.0 39.0 ! 2.0 Tc207(1) 249,500 f 2600 218,500 f 2600 104.o f 2.0 81.1 ;i 2.0 i Tc207(v) 233,150 i 2600 210,580 2600 75.7 f 2.0 109.4 f 2.0 f HTc04(s) 167,420 :l 1300 140,430 i 1300 90.5 2.0 29.9 i 2.0 HT c04(sat •) 167,760 :1: 1300 141,980 ± 1300 86.4 f 2.0 HTc04(0.3 M.) 173,300 ± 1300

0) \,.) TABLE XVIII

SUMMARY OF THERMODYNAMIC PROPERTIES OF TECHNEI'IUM AND ITS COMPOUNDS (Continued) (25°)

0 s0 so(b) Compound - 6H -6F� - 6 f f ... . .

Tc04(aq.) 173,000 i 1300 150,550 i 1300 75.3 i 1.0 45.1 i 1.0 KTc04(s) 242,600 i 1500 216,480 f 1500 80. 9 i 1.0 39.1 ± 1.0 ------NaTc04(s) 79.6 f 1.0 37.4 :l 1.0 --- NH4Tc04(s) --- lh6. J .i 1.0 43 .8 :i 1.0 --- K2TcCl 6(s) --- 117.6 :f: 1.0 79.5 f 1.0

8Five significant figures given for calculation purposes only. b1argely from Appendix V.

CX> .i::-- 85

ACIDIC SOLUTIONS

-0.782

0 -0.240 I -0.83 -0.65 Tc- ( .5) Tc --- Tc0 --- Tc 0 2 3 I -0. 472

BASIC SOLUTIONS 0. 322

0 0 88 0 0 1.00 _ Tc- ( .5) Tc . 5 Tc 0 . Tc 0 T 4 I______\_._4_7_4___3 ___ _r

FIGURE 12.- OXIDATION - REDUCTION DIAGRAM FOR TECHNETIUM 86

those of the other elements, although it resembles rhenium some­ what more closely. A summary of some of the physical and thermo­ dynamic properties of this group is given in Table XIX. The oxidation-reduction potentials for the members of sub­ group VIT. are summarized in Figure 13. The data for manganese are from a summary by Hugus and Latimer,? while the data on rhenium, with the exception of the free energy of form:1.tion of 1 rhenium trioxide, are from Latimer s tables. 8 The data on the free energy of formation of rhenium trioxide, as well as the more dependable average for perrhenate ion based on the heat of for­ mation of rhenium heptoxide, are from tnis research. The hypothetical single bond strength of the metal atom to oxygen can be calculated from the heats of formation of the oxides. Thus, the bond energy of the rhenium-oxygen bond can be obtained by dividing the heat of formation of rhenium dioxide by four (25 kcal.), the heat of formation of rhenium trioxide by six (24 kcal.), or the heat of formation of the heptoxide by fourteen (21 kcal.). The average value of 24 kcal. will be adopted. Similarly for technetium, the values are 26, 21, and 19 kcal. from the dioxide, trioxide am heptoxide, respectively. The average value of 22 kcal. will be adopted in this case. The same method is not successful in the case of manganese because the lower oxides

(7) Hugus, z., and Latimer, w. M., University of California Radiation Laboratory Report, UCRL-1088, 1951. (8) Latimer, w. M., private communication; to appear in the new edition of "Oxidation Potentials", 1952. TABLE XIX

a COMPARISON OF PROPERTIES OF SUB-GROUP Vll

Property Mn T c Re

/!, 0 b -1 b Hf KM04( s) -194.4 kcal. mole -242.6 -262.4 --- t.H f M 0 ( ) -266.1 -296.7c 2-7 s 0 c c k 1 t.F f M04( aq. ) -101.6 cal . mole- -150.6 -166�5 0 --- 1 kc 1 LiHf M03(s) - 29.0 al.mole- -146.o --- 1 d Li � HM04(s) �167.4 kcal.mole- -182.o b e LiH 1 M02(s) -124.5 kcal. mole-1 -102,500 -100.4 0 f g s250, Metal 7.59 e.u. 6.8 7.8 f --- d Dissociation Pressure o HMo4(s) at 25° 0.67 mm. 0.35 h i j Ionization Potential, M(g) 7.41 volts 7.45 7.85 Atomic Weight 54-93 98.91 186.31 1 Density, Metal 7.20 g. cc .-l(k) u.5o1 20.53 1 Metallic Radius 1.291 A.m 1.3581 1.373 (k) k M. P., Metal 126oo 214011 3167 .,,

TABLE XIX

COMPARISON OF PROPERTIES OF SUB-GROUP VIIa (Continued)

Property Mn Tc Re

(0 M.P ., M207(s) -200 ) 120 300d x Abundance (per 10,000 Si atoms)P 77 present .0041. 6 q r Magnetic Susceptibi lity Me tal x 10 c.g.s. units 27 27o 6 q 5 (er) 9 8 Solubility of M2S7 in 9 !i.!_ HC1 s. s. insoluble o t U X V Ks.p.' KM04(s) o:66 c2o ) 0.44 (270) 4. 4 10-J (JOO) M - 0 Bond Energy 30 kc al . ?? 24

w Stability of Aqueous Hexavalent State unstable unstable stable

aAll technetium values from this research unless otherwise indicated . bnselected Values of Chemic al The rmodynamic Properties" , National Bureau of S tandards , Wa shington , D. C. , 1948- 1950, Series I, II, III. '1iecalculated from (h) and values fro m Appendix IV.

dsmi th, T. , Jr. , Line , L. , and Bell , A. , � Am. Chem. Soc. , in �ress (1952). w. w. co co ecalculated from ref. c and data of Hugus, z, and La timer , w. H., to appear in the new edition of "Oxidation Potentials", 1952. TABLE XIX

a COMPARISON OF PROPERTIES OF SUB-GROUP VII (Continued)

fKelley, K. K., Bur . Mines Bull ., 477_ {1�50) �

gEstimated by the method given in Chapter VI.

hBacher, R. F., and Goudsmit , s., "Atomic Energy State s" , McGraw-Hill, New York, N. Y., 1932, p. 278.

iMeggers, F. , J. Resear-cn 1'J a,:,l. Bur . Standards , 47 , 7 (1951) . w. . - .-- - jMeggers , w. F., ibid., �, 1027 (1931) .

kHodgman ; C. D. , "Handbook of Chemistry and Physics" , Chemical Rubber Co., Cleveland, Ohio, Twenty-Ninth Ed ., 1945 , p. 409 ,

1Mooney, Rose C. L. , Acta. Cryst., .!, 161 (1948) .

ITlpauling, Linus , "The Nature of the Chemical Bond" , Cornell Univers1ty Press, Ithaca, N. Y., 1945, p. 409 . nParker, G. and Martin, J. Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-1260 (1w.951, \ . w.,

°Mellor, J. w., "Treatise on Inorganic and Theoretical Chemistry" , Longmans , Green and Co., New York, N. Y., Vol. 12, 1932, p. 466.

PBrown, Harrison, � Mod • . Phys. J. 21, 625 (1949).

, \,() TA BLE XIX

a COMPARISON CF PROPERTIES OF SUB-GROUP VIl (Continued)

5 Perrier, C. ' and Segre ' E. ' J. Chem. Phys. ' ]_, 155 ( 19 39) • tEstimated from data in Mellor� J. ·w. ; · 11Treatis·e · on Inorganic and Theoretical Chemistry", Longmans, Green and Co., New York, N. Y., Vol. 12, 1932, p. 466. uParker, G. w., Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-1116 (1951). vGoldin, A. S., 11A Radioactive Tracer Investigation of Rhenium Chemistry", Doctoral. Dissertation, The University of Tennessee, 1951. Wsee Figure 13. XMerrill, P. w., Science, 115, 484 (1952) .

'D 0 91

-0.115 -1.695

+2 Mn 1. 18 M n - 1.229 1 1 -2.26 ; -0.564 Mn02 Mno Mno; - 0.781 I -0.782

{ 0.5) -0.240 -0.83 -0.65 - Tc- Tc Tc02 Tc03 Tc04 I -0.472 I

-0.510

0. ) -0.252 -0.38 -0.76 Re-( 4 Re Re02 Re03 I -0. 361

ACID SOLUTIONS

FIGURE 13.- OXIDATION - REDUCTION DIAGRAM FOR SU B-GROUP :sz:rr 92

have added stability over the higher oxides by reason of the partial ionic character of the o::xygen-metal bonds. Thus the single bond energy for manganese(II) oxide is 46 kcal ., while manganese(IV) oxide gives 31 kcal ., and the manganese oxide, Mn304, gives 28 kcal . In this case, the average values from the latter two oxides will be adopted (30 kcal .} since the lower oxides of technetium and rhenium have not yet been prepared. Since technetium resembles rhenium in many of its properties, the following differences are, perhaps, of more interest: (1) The ionization potentials of the gaseous ions indicate that technetium is in this respect more like manganese. This is undoubtedly due to the fact that the coulombic attractive force, which is dependent up on the square of the charge, is of prime importance, the screening effect of the extra-nuclear electrons being secondary. (2) Technetium hep toxide differs markedly from rhenium hep­ toxide. However, there is not enough information available on manga­ nese heptoxide to know how technetium heptoxide resembles it, if at all. The difference between the heptoxides of rhenium and technetium undoubtedly is due to some effect in the rotational entropies of these compounds and their electrical properties near the melting points. The data seem to indicate that technetium heptoxide under­ goes a type of pre-melting at low temperatures, although confirmation of this hypothesis llll.lst await a more complete study of the crystal 93 structure data.

(3) '!'he data on the oxidation potential s of the rhen ium trioxide couples and the estimates on the technetium trioxide couples indicate that while these two compounds are sonewhat similar, rhenium tr ioxide is st able wit h respect to disproportiona­ tion while manganese(VI ) and technetiu m trioxide are unstable in this respect. (4) The and melting points of these elements are different. (5) The solubilities of the potassium salts of the heptavalent states are different, potassium pertechnetate being the more soluble.

(6) Perrhenic and pertechnic ac ids, while similar in that both are unst able with respect to loss of water vapor, differ in that pertechnic acid has a relat ively higher dissociation pressure than perrhenic acid at normal temperatures. (7) The elements of su b-group VII are, of course, quite different in their abu ndance in the crust of the earth. No st able isotope of technetium has yet been discovered; Tc99 with which this research was carried out has a half-life of 2 x 1o5 years.

Bot h rhenium and manganese have stable isotopes. CHAPTER VII

SUMMARY The new element, technetium, has been studied thermodyna­ mically to provide data which are useful in comparing its pro­ perties with those of its neighbors in the periodic table. The free energies and enthalpies of formation have been determined or estimated by methods of combustion calorimetry and potentiometry, and an oxidation-reduction sche:ioo for the element has been formulated. The thermodynamic measurements were preceeded by the preparation, purification and identification of the following compounds: (1) technetium heptoxide, (2) ammonium pertechnetate, (3) technetium dioxide, (4) pertechnic acid, and (5) technetium metal. In addition, thermodynamic infonnation has been estimated for technetium trioxide from observations on its chemistry. With this chemical thermodynamic information, similarities and differences between technetium and its compounds and other members of sub-group VII have been pointed out. In general, tech­ netium has been found to resemble rhenium more cl ose]y th an manga­ ne se, although a number of exceptions to this generalization have been pointed out. The properties of this new element have again emphasized the usefulness of the periodic arrangement of elemen ts. Additional thermodynamic determinations have also been made on the heat of for mation of rhenium heptoxide and rhenium trioxide 95 to supplerTBnt the data already exist:ing in the literature on these compounds. APPENDIXES APPENDIX I

X-RAY DIFFRACTION POWDER PATTERNS

X-ray powder diffraction patterns have proved of value in th is research to demonstrate (1) whether the sanB compound may be produced by different procedures, and (2) whether analogous technetium and 1 rhenium compounds are isomor phous.

A. Technetium and Rhenium Heptoxides

A compari son of the interplanar distances found in rhenium heptoxide and te chnetium heptoxide is afforded by Table XX. It can be seen that these two compounds are not isomorphous. The patterns are not being classified at this time and are not comparable to any other known oxide structure.

B. Technetium Metal

2 Mooney has determined the powder pattern for technetium me tal prepared by reduction of technetium sulfide with hydrogen. The metal prepared by hydrogen reduction of ammonium pertechnetate in this research gives the same pattern (s �e Table x.n) .

(1) The author is indebted to Mr. Ray Ellison, ,chemi stry Division, Oak Ridge National Laboratory, for his determinations of all of the X-ray powder patterns reported in this dissertation. Details on these struc­ ture s will be published by Mr. Ellison at a future date. (2) Mooney, Rose c. L., Ac ta. Cryst., 1, 161 (1948). 98 TABLE XX

COMPARISON OF X-RAY DIFFRACTION PATTERNS FCR TECHNETIUM AND RHENIUM HEPI'OilDES

a. Tc207 Re2o7a.

d(A)b Intensity d(A)b Intensityc

4.2 w

a.Pattern determined by Mr. Ray Ellison , Chemistry Division, Oak Ridge National Laboratory.

binterplana.r distance in angstroms. CSample shows preferred orientation and accura cy of determination is poorer for certain lines.

dsymbols : �, strong; �, nedium; ,!, weak. 99 TABLE XXI

COMPARISON OF TECHNETIUM METAL X-RAY PATTERNS

Mooneya This Researchb

Sin2e Intensityc Sin2e Intensity

0.1065 m- 0.1061 m 0.1241 m- 0. 1234 mi 0.1376 vs 0.1363 vs+ 0.2306 w 0.2304 wt 0.3182 m- 0.3171 m 0.3829 mt 0.3826 m o.4i37 vvw 004235 vw- Oo440 6 s- o.4393 m 0.4533 m 0.4531 m- 0.4930 vvw 0.4907 vvwd o.5465 vw� Oo5445 vvwd 0.5980 vw Oo5947 vvwd 0.6984 m- 0.6978 m- 0.7377 vw 007365 w- 0.7687 s 0. 7681 mf 0.8070 m 0.8068 m 0.8604 m- 0.8603 w- 0.8727 m-+ 0. 8729 w 0.9124 w 0.9481 mi

aMooney , R. c.1., Acta Cryst., 13 161 (1948); metal prep ared by sulfide reduction: bpattern determine d by Mr. Ray Ellison, Chemistry Di vision, Oak Ridge National Laboratory; metal prepared by hydrogen reduction of ammonium per­ technetate. cSymbols � �, strong ;,!!!, medium; �, weak; !_, very.

dDu e to faintness and location of these lines on film , poorer agreement in data so indicated is to be expected. 100

C. Technetium Dioxide

A comparison of the diffraction patterns for technetium dioxide prepared by pyrolysis of ammonium pertechnetate, (A), with the dioxide prepared by dehydrating at high temperatures the hydrate formed from the solution reduction of ammonium pertechnetate by zinc and hydro­ chloric acid, (E), is given in Figure 14. Interplanar distances are given in Table XXII for dioxide (A ). It can be seen from Figure 14 that the same compound results in both cases. The hydrated dioxide has a different pattern. The distances in Table XXII do not coincide with those repor ted by Zachariasen3 for a dioxide prepared by pyrolysis of ammonium pertechnetate.

(3) Zachariasen, W. H., Argonne National Laboratory Report ANL-4082, 1947 ; to be published in Acta. Cryst. TABLE XXII 101

INTERPLANAR DISTANCES FOR TECHNETIUM DIOXIDEa PREPARED BY PYROLYSIS OF AMMONIUM PERTECHNETATE

d(A. )b Intensityc

3.363 m 3.316 vs 2.425 s 2.352 w+ 2.158 w 2.109 vvw 1.874 vw 1.852 vw 1.792 vw 1.696 s- 1.668 m- 1.539 w- 1.499 wt 1.490 wf 1.461 vw 1.420 w 1.382 w+ 1.366 tr 1.309 wf 1.219 w- 1.192 wt 1.171 vw

anetermined by Mr. Ray Ellison, Chemistry Division, Oak Ridge National Laboratory. brnterplanar distance in angstroms. csymbols: s, strong; �, medium; !, weak; :!, very; tr, trace. A

B

FIGURE 14.- X-RAY POWDER PATT ERNS FOR TECHNETIUM DIOXIDE SAMPLES APPENDIX II

PHYS ICAL CONSTANTS

Various physical constants have been used in th is research in the calculation of thermochemical quantities. Table XXIII lists the values used in all the calculations. The atomic weight of technetium has not yet been determined chemically with a high enough precision to be used as a standard; the only chemical value reported is the one 1 from this research: 98.8 i O.l. The calculated theoretical iso- 2 topic rra.ss, assuming only one isotope, Tc99, is 98.95. The mass

99 3 spectrometer value for Tc mass is 98.913. Since there is no evidence at present 4 for any other long-lived isotopes of technetium formed by uranium fission, the mass spectrometer value of 98 .91 will be used. None of the mass dependent val ues reported in this research is known to better than about one percent, so the differences in values given above ar e not significant.

(1) Cobble, J. w., Nelson, c. M., Smith, w. T., Jr., and Boyd, G. E., �Am. Chem . Soc., 74, 1852 (7q�2). ( 2) Metropolis , N. , and Reitwi esner, G. , U. S. Atomic Ener gy Commission Document, NP-1980, 1 9�0. (3) Inghram, M. G., Hess, D. C. , Jr ., and Hayden, R. J., Phys. Rev., 72, 1269 (1947 ). (4) Boyd, G. E., Record Chem. Progress (Kresge-Hooker Sci . LiQ. ), Spring, 1951, p. 67 . 104 TABLE XXIII

PHYSICAL CONSTANTS

Defined calorie = 4. 181.t) absolute joulesa = 4.1833 int . joulesa Gas. constant, !! = 1.987 defined caloriesb Ice point, °K. = 273.16c ln N = 2.303 log N Atomic weight of Tc = 98.91d 0 International volt equivalent = 23,061 defined calories

Atomic weights of other elements·e

8aossini, F. n., � Research � Bur. Standards, §., 1 (1931). busing R = 8.31436 x 107 ergs mole-1 from Bearden, J. A., and Watts, H. M., � Rev., 81, 73 (1951) and the defined calorie = 4.18�lo'Tergs as given above. 0 Birge, R. T., Rev. Mod. �, 12, 233 (1941) . drnghram, M. G., Hess, D. c., Jr., and Hayden, R. J., Phys. Rev., 72, 1269 (1947). eCommittee on Atomic Weights, .:I.!_ Am. Chem. Soc ., 72, 1432 (1950). APPENDIX III

TREATMENT OF THE CALORIMETRIC DATA

Figure 15 gives a typical curve ob tained for the change in resistance of a platinum resistance thennorooter with time when technetium was burned in the combustion calorimeter previously des­ cribed (Ch apter IV). Since the resistance thermometer resp onds linearly to the temperature over the one degree range illustrated, Figure 15 may also be considered a time-temperature curve. The curve illustrates the salient features of tilll3-temp era­ ture curves for such calorimeters : the initial, small, linear ri se corresp onds to a heat leak into the calorimeter from the bath. At the ignition time, the contents of the bomb begin to burn and supply he at to the calorimeter and the temperature then rises quite rapidly. This "rise curve" is almost exponential after the first few se conds .

It then reaches a maximum value, an d, dep ending upon whether the final temperature inside the calorimeter exceeds or is lower than that of the bath, the temperature falls or continues to rise at a small, linear rate.

The problem is , what part of the temperature change should be attributed to the heat of combustion and what part to radiation leakage? One simple method is to take, at some arbitrary time, the difference between the extrapolations of the initial and final linear portions of the curve. However, the net rise will then dep endup on the time at which it is computed, since, in general, the slopes of the two linear portions will not be the same. If the 11rise curve" 0.09 ------r------1 -i- .- r-

0.08

IGNITION TIME 0.6 RISE TIME 0.07 ' I ·-·--·-·-·-· .

0.06

0.05

-9. 0.04 a:::

0.03 0.01 ohm z 0.1 degree

0.02

0.01

--·---� o• -· •·- O 500 1000 1500 2000 2500 TIME (sec.)

I-' 0 FIGURE 15.- TIME-TEMPERATU RE CURVE FOR CALORIMETER 107 were symmetrically shaped, the net temperature change could be taken at a time corresponding to the inflection point.

It has been found experimentally by Dickinson1, however, that most of the calorimeter "rise curves" are exponential after the first few seconcsafter firing. When this is the case it is recommended that the net temperature change be taken at a time such as to give equal areas under the rise-time curve (Figure 15). For an ideal exponential curve this would correspond to 0.63 times the total net rise, but for most practical calorimeters, it co?Tesponds to about 0.60 times the net rise. This time can be determined either empirically or analy­ tically. The 0.6 rise-tine method was used for the calculation of all net calorimetric temperature changes in both the combustion and solution calorine ters.

(1) Dickinson, H. c., U. � Bur. Standards Bull., 11, 189 (1914). APPENDll IV

THE HEATS CF COMBUSTION OF RHENIUM AND RHENIUM TRIOXIDE

To test the accuracy of the combustion calorimeter rhenium metal was burned as described previously (Chapter IV ). The heat of combustion of rhenium in o:xygen has been determined by Roth and 1 Becker. Since technetium and rhenium are similar, the latter appeared to be the best available secondary standard to test the calorimetric equipment. The heat of combustion of rhenium trioxide to form rhenium heptoxide was also measured, since the only data previously available on this compound were reported by Roth and Becker who used indirect methods. More accurate heat data on rhenium trioxide were desired to assist in estimating some of the thermodynamic properties of technetium trioxide, which has not yet been prepared in large enough quantities to burn.

A. Heat of Combustion of Rhenium Metal

Rhenium metal,2 prepared by hydrogen reduction of ammonium perrhenate at 5000, was used directly. It was fo und that rhenium burned somewhat more easily than technetium metal, and because of its higher density and compactness, less oil accelerator was required.

(1) Roth, w. A. , and Becker, G., � physik. Chem. , A 159, 27 (1932); data has been recalculated and s1.llllm:lrized in "Selected Values of Chemical Thermodynamic Properties", u. s. Bureau of Standards, Washington, D. c., 1948, Series I. (2) Kindly furnished by the Rhenium Plant, The University of Tennessee. 109

The data obtained using about 100 mg. samples are given in Table XXIV. Taking 186.31 as the atomic weight of rhenium, and 1397.4 cal .deg.-l as the heat capacity for the calorimeter (as given in Chapter IV), the heat of combustion estimated from Table XXIV becomes -149 .J i 0.2 kcal. mole-1 of Re, or -298.6 ± 0.4 kcal. mole-1 of Re207. The process �asured was :

2Re(s) f 7/2 02( g. ,30 atm. ) f H20(l) t aq. = 2HRe04(Fr.) (A IV-1) /:::.E : -298.6 ± 0.4 kcal. where M. represents the average perrhenic acid concentration in the resulting bomb soluti on. Due to the lower volatility of rhenium heptoxide compared with technetium heptoxide, only about one-half of the products of the rhenium coni:>ustion went into the bomb solution. The remainder was found in the ignition cup as fused heptoxide. Be­ cause of the relatively low saturation pressure of water vapor in the gas at 30 atm., the heptoxide apparentzy did not absorb water dlll'ing the run. As with technetium, corrections to: (1) standard state con­ ditions of unit fugacity for the oxygen and for (2) change of the measured /:::. E into the corresponding /j,,H of combustion must be made , (See Chapter IV):

2Re(s) 7/2 0 + H 0 + aq. = 2HRe0 t 2(g ) 2 (1) 4(M . ) (A IV-2) �H = -300.9 i o.4 kcal . TABLE XXIV

HEAT OF COMBUSTION OF RHENIUM METAL

a 6 Tobs. 6Tcorr. Wt. oil 6.Toil 6.Tnet Wt . Re Temp. Rise g.-1 . .

o.65610 o.6514° 73.01 mg. o.5775° 0.0739. . -- ° 129.09 mg. o.573o g.-1 0.4511 o.4464 49 .24 0.3895 0.0569 98.77 0.576 0.6214 0.6167 69.93 0.5532 0.0635 111.00 0.572

Average : 0.574 ! 0.001° g.-1

a.corrected for fuse ignition and nitric acid formation.

g 111 The heat of combustion of the metal to form aneydrous rhenium heptoxide was calculated on the basis that half of the heptoxide dissolvErl in the water in the bomb to form a 0.1 M. solution of perrhenic acid. Using 5 kcal . for the heat of solution of a half-mole of rhenium hep­ toxide to form a 0.1 l1:, perrhenic acid solution3 the following is obtained:

2Re(s) + 7/2 02(�. ) = Re207(s) (A IV-3) .6H : -295 .9 ¼ 2.0 kcal. mole-1

Any error is largely due to the uncertainty about the final state of rhenium heptoxide . This value may be compared with that given by the Bureau of Standards4 of -297.5 i 2.0 kcal. mole-1 • The agreement is considered satisfactory; for subsequent calculations the average value, -296.7 i 0.8 kcal. mole-1, will be used.

B. The Heat of Combustion of Rhenium Trioxide

Rhenium trioxide5 was also burned in the sa.re manner as was rhenium metal. Very little heat was given off (Table XXV) . In fact, after the 11 oil heat11 was subtracted, the remaining temperature rise was barely outside the experimental errors involved, indicating that

(3) Estina.ted from the data given by Roth, w. A., and Becker, G., !!_ physick. Chem. A 159 , 27 (1932) . (4) "Selected Values of Chemical Thermodynamic Properties", u. s. National Bureau of Standards, Washington, D. c., 1948, Series I. (5) Kindly furnished by the Rhenium Plant, The University of Tennessee, Knoxville, Tennessee. TABLE XXV

THE HEAT OF COMBUSTION OF RHENIUM TRIOXIDE

.. b 1 6Tobs. 6T�orr. Wt . oil 6Toil 6 Tnet Wt . Re03 Temp. rise g. -

° ° ° -1 0.9267.. 0.92200 116.. �26 mg. 0.9196 0.0024 253�1: mg. 0.0095° g.

1.1438 1.1391 142.97 1.1309 0.0082 253.7�- ... 0.0323 0.6245 0.6198 78.26 0.6190 0.0008 78.3 0.0103

Average: 0.0174 :l 0.0080

�orrected for fuse ignition and nitric acid formation.

Drhis represents·-about 80�90· percent 6£ the· cnarge; t.he- ainount actually bur ned was obtained from the loss in weight of the combustion sample. ll3

the difference between the heats of formation for the trioxide and heptoxide is very small per ·rhenium atom. The fraction of rhenium trioxide burned, which was det ermined by loss in weight of the rhenium trioxide sample amounted to about 80-90 percent. The process measured was : •

(A IV-4)

�El : -11.4 :1: 6.0 kcal.

Taking the value of 0.0174 deg. g.-1 from Table XXV (4.08 deg. mole-1 and of Re03), using the value of 1389.l cal. deg.-1 for the heat capacity of the calorimeter (see Chapter IV) the heat of reaction for (A IV-4) becomes -5.7 ± 3.0 kcal. mole-1 of Reo3 or -ll.4 f 6.0 kcal. mole-1 of Re207 formed. As before, it was found that only half of the rhenium heptoxide formed went into solution. After making the corrections for the standard state of the oxygen, changing from lE to � H, and subtracting 5 kcal. for the heat of solution of half a mole of rhenium heptoxide, th e follow­ ing was obtained:

2Re03(s) + �2( g. ) _= Re207(s) (A IV-5) � H5 = -6 f 6 kcal.

The heat of formation of rhenium trioxide may now be calculated from its heat of combustion, and the heat of fornation of rhenium heptoxide, -296.7 kcal. mole-1: ll4

(A IV-6)

Roth and Becker obtained -83 kcal. mole-1 for this heat of formation. The poor agreement is due to the fact that their calcu­ lation was based upon the differences observed when rhenium was burned to form a mixture containing the heptoxide and varying amounts (4-8 per­ cent) of the trioxide. The direct determination reported here is con­ sidered more accurate. APPENDIX V

ESTIMATION OF ENTROPY VALUES FOR TECHNETIUM AND ITS COMPOUNDS

The heat and cell EMF data obtained in this research can be made more useful by the estimation of suitable entropy values for tech­ netium and its compounds. Thus, by the relation:

(A V-1) free energy data can be estimated from calorimetric heats of reaction, and the heats of formation and reaction can be calculated in turn from cell voltages (free energy changes). These estimations are utilized in the calculation and estimation of many thermodynamic quantities given in Chapter VI.

A. The Entropy of Technetium Me.tal

The entropy of technetium metal can be estimated rather accurately from its position in the periodic table. Plotting the entropies1 of the transition elements at 25° against their atomic numbers, it was observed that each period formed a parabolic shaped curve, with man­ ganese, technetium and rhenium forming the vertex for each period. The entropy of technetium metal was estimated to be 6.8 cal . mole-1 deg.-l (e .u.) by this empirical method. This may be compared with the value estimated by Brewer2 of 8 . e.u.

(1) Kelley, K. K., Q:_§.:.�of Mines Bull. 477, 1950. (2 ) Brewer, Leo, Paper 3 in 11Chemistiy and Metallurgy of Mis­ cellaneous Materials : Thermodynamics", edited by L. L. Quill, McGraw-Hill, New York, N. Y., 1950. U6

B. The Entropy of Tc04(aq.)

Latimer3 has shown that the entropies of simple ions (assuming

Htaq.) = 0) can be represented by the equation

0 2 S = 3/ 2 R ln M + 37 - 270 Z/re (A V-2)

where Mis the mass of the ion in atomic weight units, Z is its

charge in e.s.u., and re is the effectiv e ionic radius. For simple ions, r e can be calculated from the ionic crystal radii. It is known that perchlorate, perrhenate and pertechnetate ions are isomorphous and that their ions can be coprecipitated from solution. If it is

2 - assumed that equation (A V- ) also holds for xo4 ions, and that Z/r� - - - is the same for Tc04 , Reo4 and Cl04 , then on using the value fo r the - 4 of entropy of the c104 ion as 43 .6 f 0.5 e.u., the entropies Tc04(aq.) and Re04(aq .) are calculated to be 45.1and 46.4 e.u., respectively.

c. The Entropy of KTc04(s)

Knowing the value for the entropy of the Tc04(aq.) ion the entropy of KTc04(s ) may be calculated using the solubility measure­ 5 ° ments of Parker. Using his value for the Ks.p. at 7 of 0.14 and

(3) Powell, R. E. , and Latiner, w. M., J. Chem. Phys., 19 , 1139 (1951) 0

(4) Kelley, K. K., Q.:. §.:_ Bur. of Mines Bull., 477, 1950. (5) Parker, G. w., Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-lll6, 1951; see also Table XVII. 117

° 0.44 at 27 , the heat of solution is 9550 cal. mole-1, arrl LiFat

° 0 27 is -RI' 1n Ks.p. = 490 cal. mole-1. AS can then be estimated to be 30.2 e.u. for the process: + Tc0 aq. + KTc04(s) : K(aq.) t 4(aq.) (A V-3)

0 6.s .:: 30.2

As is usually done for this type of calculation, corrections for dilution of the saturated solution to infinite dilution have been neglected, as well as the change from 27° to 25°. Taking the Kt (aq.) 0 . entropy as 24.2 e.u. then s for KTc04(s) becomes J9.l e.u The entropy of KTc04(s) was also estimated from the entropy

? of KC104(s) using the equation proposed by Latimer for similar substances :

0 at. wt. - s 3/2 R 1n _____,_...___._'-- (Cl) (A V-4) �Cl0 KTc0 = 4 4 at. wt.(Tc)

(8) Using the value of J6.l e.u. for the entropy of KC104(s) , the en tropy of KTc04(s) calculated from equation (A V-4) was again 39.1 e.u.

(6) Kelley, K.K., .!!.:_§.:.�Mines Bull., 477, 1950. (7) Latimer, W. M., "Oxidation Potentials", Prentice-Hall Pub. Co., New York, N. Y., 1938, p. 328. (8) Kelley, K. K., loc. cit. 118

D. Entropies of Solid Technetium Compounds of the Formula Ml'c04

Latimer 9 has mownthat the entropies of solid ionic compounds may be assigned in an additive manner to the cation and anion. The contributions of all of the common ions have been averaged into con­ sistent values and, hence, the entropy of many solid compounds can be estimated with fair accuracy. Thus, the entropy of KC104(s) is 0 0 equal to s + + s - = 9.2 + 26.0 = 35.2 e.u. which may be K (s) c104(s) 10 compared to the experimental value of 36.1 e.u. • Using 39 .1 e.u. 11 farKTc0 4(s) as given above, and 9.2 e.u. for K1s) , the calculated contribution of Tc0 in solid salts is 29.9 e.u . Using 13.9 e.u. 4(s) 12 for NHt(s), 0.0 for Hts) and 7.4 for Nats) from Latimar's table, ° then the entropies of NH4Tc04(s), Hl'c04(s) and NaTc04(s) at 25 are 43.8, 29.9, and 37-4 e.u. respectively.

E. The Entropy of K2TcCl6

The entropy of potassium hexachlorotechnetate(IV) may be cal­ culated by comparing it with the general class of salts of the type

K2MCl6, arxi breaking up the total entropy into that of the component l3 elements. Using 18. 4 for K2 , 48.6 for Cl6 and 12.5 for technetium,

Am. (9) Latimer, w. M. , � Chem. Soc., 73, 1480 (19.51). (10) Kelley, K. K., JL.. _h Bur. Mines Bull., 477, 19.50. (11) Latimer, w. M., � cit. (12 ) Latimer, w. M., loc. cit. (13) Latimer, W. M. , loc. cit. ll9 the entropy of K2TcCl6 is 79.5 e.u. at 25°. The value of 12.5 e. u. for the contribution of technetium in its compounds is based upon 1 Latimer •s 4 empirical curve for me��llic atoms in compounds, and not upon the absolute entropy of technetium (6.8 e. u.).

F. Entropies of Technetium Oxides

Using the sane additivity principles, the entropies of oxides may be calculated with some accuracy. In this case, however, the contribution by the oxygen atoms to the total entropy depends upon the coordination number of the metallic constituent , sim e the bond distances and stabilities of the various oxides of the same element will be, of course, somewhat different. The average contributions for oxygen in the oxides of the type M02, M03 and M207 have been estimated from oxides for which entropy data are available.15 Thus, the entropy of Tc02 becomes 12.5 4 2(1. 2) or 14.9 e. u., Tc03 is 12.5 + 3(1.l) or 15.7 e.u. , and Tc207 (calculated as twice Tc03.5) is 2(12.5) + 7(2.65) or 43.5 e.u. This latter value may be con­ siderably in error because of the paucity of data for the higher oxides. The entropy of fusion (see Chapter III) for this compound indicates that it may have an abnormally low entropy at room tempera­ ture. Therefore, the value calculated from equation (VI-12) and

(14) Latimer, w. M. , :!..!_ Arn. Chem. Soc., 73, 1480 (1951) . (15) Kelley, K. K., --U. S. --Bur. Mines Bull. , -477, 1950, 120 based upon the entropy of HTc04(s) will be used: 39.0 e.u.

The Entropy of Tc G, (g)

The entropy of gaseous technetium at 25° has been calculated

16 from the equation:

o so - s o + s - trans. e (A V-4) where S�rans. is the translation entropy for a monatomic gas and is 0 given by the Sakur equation as 3/2 R ln M � 25.996. s is the e entropy associated with the electronic llDlltiplicity of the ground

1 o o o 6 ; 1 7 state f the technetium at m. This gr und state is a s5 2 leve ; thus the multiplicity, p = 2j f 1, is (2) 5/2 t 1 or 6. S� is 0 equal to R ln p, or 3.561 e.u. o strans. has the value f 39.694; therefore, the total entropy at 25° is 43.26 f 0.01 e.u. The other energy levels for the atom are too far above the ground state ( the lowest is 2572.9 cm.-1 above the ground state) to contribute to the entropy at room temperatures.

The various entropies calculated and estinated for technetium and its compounds are summarized in Table XXVI. For other entropies based upon measurements performed in this research but also involving the entropies estimated in this appendix, see Chapter VI.

(16) Kelley, K. K., .!I.! S. � Mire s Bull., 477, 1950. (17) Meggers, W. F., !!.!_ Research Nat. � Standards, 47, 7 (1951). 121

TABLE XXVI

SUMMARY OF ENTROPlES ESTIMATED FOR TECHNETIUM AND SOME OF TI'S COMPOUNDSa

Compound S0 250

Tc b ( g) 43.26 ± 0.01 e.u. ± Tc(s) 6.8 0.2 Tco4- ( aq.) 45.1 ± o.4

KTc04(s) 39 .1 ± 1.0 NH4Tc04(s) 43. 8 ± 1. 0

HTc04(s) 29.9 ± 2.0

NaTc04(s) 37 .4 i 1.0

K2TcCl6(s) 79.5 f 1.0 c 4 T 02(s) 1 .9 i 0.5 Tc03(s) 15. 7 .i o.6 c T c207 (s) 39.0 i 2.o

asee also T able XVIII.

�rrors estim ated from deviations observed using the same nethod of estimation for compounds of lmown entropies. Crhe error for this compound may be larger because of the abnormality observed in its entropy of fusion (See Ch apter III ). BIBLIOGRAPHY BIBLIOGRAPHY

Bacher, R. F. and Goudsmit, s., 11Atomic Energy States", McGraw-Hill, New York, N. Y., 1932, p. 278. Bearden, J. A. and Watts, H. M., Phys.�, 81, 73 (1951). Birge, R. T.,� Mod. Phys., 12, 233 (1941). Boyd, G. E., Record Chem. Progress (Kresge-Hooker � Lib.), Spring, 1951, p. 67. w., c. w. Boyd, G. E., Cobble, J. Nelson, M., and Smith, T., Jr., -J. Am. Chem. Soc. , 74, 556 (1952).

11 Brewer, Leo ., Paper 3 in Chemistry and Metallurgy of Miscellaneous Materials: Thermodynamics", edited by L. L. Quill, McGraw-Hill, New York, N. Y., 1950.

Brown , Harrison, Rev. Mod. Phys., 21, 625 (1949). Cobble, J. w., Nelson, c. M., Parker, G. w., Smith, w. T. , Jr., and Boyd, G. E. , i!..:.. �Chem. Soc., 74, 1852 (1952) . Committee on Atomic Weights, --J. Am. Chem.' --Soc., -72, 1432 (1950). Committee on New Elements, International Union of Chemistry, Chem. Eng. News, 27, 2996 (1949). Dickinson, H. c., Natl. Bur. Standards Bull. , 11, 189 (19l4j. 11 1 Druce, J. G. F. , Rhenium 1 , The University Press, Cambridge, England, 1948, p. 32. Fried, S. , � Am. Chem. Soc., 70, 442 (1948). Fried, s., Private Communication to G. E. Boyd, 1950 .

Geilrnan, w. and Wrigge, F. w., -z. anorg. -u. allgem. Chem. , 222, 56 (1935). 1 Goldin, A. s., 1 A Radioactive Tracer Investigation of Rhenium Chemistry", Doctoral Dissertation, The University of Tennessee, 1951, Grove-Rasmussen, K. v., Acta. Chem. Scand., 2, 422 (1951). Hackney, J. c. , iL:.. Chem. Ed. , 28, 186 (1951). Harned, H. S. and Ehlers, R. w., �Am. Chem. Soc. , 54, 1350 (1932). 124 Harned, H. s. and Owen, B. B., "The Physical Chemistry of Electro­ lytic Solutions", Reinhold Publishing Company, New York, N. Y., 1943, PP• 541, 547, 119 . Hodgman, c. D., "Handbook of Chemistry and Physics", The Chemical Rubber Company, Cleveland, Ohio, Twenty-ninth Edition, 1945, p. 409.

Hugus, z. and Latimer,11 w. M., To appear in new edition of "Oxidation Potentials , 1952 . Hugus, z. and Latimer, w. M., University of California Radiation Labo­ ratory Chemistry Division Quarterly Report, UCRL-1196 (1951) . Inghram, M. G., Hess, D. c., Jr., and Hayden, R. J., Phys. �, 72, 1269 (1947 ). Kelley, K. K., u. s. Bur . Mines Bull ., 393 (1936); 383, (1935) ; 477, (1950) . ------Kolthoff, I. M. and Furman, N. H., 11Potentiometric Titrations", John Wiley and Sons, New York, N. Y., 1931, pp. 45, 280. Latimer, w. M., i!..!_ Am. Chem. �, 73, 1480 (1951). Latimer, w. M., "Oxidation Potentials", Prentice-Hall, New York, N. Y., 1938, P• 277° Latimer, w. M., Private Communication ; to appear in the new edition of "Oxidation Potentials", 1952. Latimer, w. M. and Hildebrand, J. H., "Reference Book of Inorganic Chemistry", MacMillan Company, New York, N. Y., 1943, p. 374. 1 11 Macinnes, D. A., 1The Principles of Electrochemistry , Reinhold Pub­ lishing Company, New York, N. Y., 1939, p. 254. Meggers, w. F., J. Research -Natl . -Bur. Standards, -6, 1027 (1931) ; -47, l (1951) .- Melaven, A. D., Fowle, J. N., Brickell, w. and Hiskey, c. F., "Inorganic Synthe sis", Vol. III, McGraw-Hill, New York, N. Y., 1950, pp . 187, 188. Mellor, J. w., "Treatise on Inorganic and Theoretical Chemistry", Vol. 12, Longmans, Green and Company, New York, N . Y., 1932, p. 466 . Merrill, P. w., Science, 115, 484 (1952) . Metropolis, N. and Reitwiener, G., U. s. Atomic Energy Commission Document, NP-1980 (1950) . 125 Mooney, Rose c. L., Acta. Cryst., _!, 161 (1948) ; Phys. �, 72, 1269 (1947) . Motta, E. E., Boyn, G. E. and Larson, Q. v., Mon-c-169 (1946) ; Phys . Rev., 72, 1270 (1947). National Bureau of Standards, "Selected Values of Chemical Thermo­ dynamic Properties", 1948-1950, Series I, II, and III .

Nechamkin, H., Kurtz, A. N. , and Hiskey, c. F., -J. --Am. Chem. Soc. , 73, 2828 (1951) . Nelson, c. M., 11The Magnetochemistry of Technetium and Rhenium" , Doctoral Dissertation, The University of Tennessee, 1952. Noddack, w., Tacke, I. and Be rg, O, Naturwissenschaften, 13 , 567 (1925) . Pa rker, G. w., Oak Ridge National Laboratory Chemistry Division Quarterly Report, ORNL-1116, 1951. Parker, G. W. and Martin, W. J • , Oak Ridge National Laboratory Chemistry Division Quarte rly Repo rt, ORNL-1260 (1951). Parker, G. w., Reed, J., and Ruch, J. w., AECD-2043 (1948). Pauling, Linus, "The Nature of the Chemical Bond" , Cornell University Press, Ithaca, N. Y., 1945, p. 409. Perrier, c. and Segr�, E., !!.:_Chem. Phys ., 2,, 712 (1937); l, 155 (1939) . Phipps, T. E., Spealman, M. L. and Cooke, T. G., !!.:_Chem. Ed. , 12, 321 (1935). Powell, R. E. and Latimer, w. M., !!.:_Chem. Phys ., 19, 1139 (1951) . Prandtl, w., �, 60, 621 (1927) . Prandtl, W., � angew. Chem., 39 , 1049 (1926); 40, 561 (1927). Rossini, F. D., -J. ----Research Natl�--- --Bur. ----Standards, -6, 1 (1931). Rossini, F. n. and Deming, w. E., !L:_ Wash. Acad. Sci., 29, 416 (1939).

Rossini, F. D. and Frandsen, M., J. Research Natl. Bur. Standards, -9, 733 (1932) . - - Roth, w. A. and Becker, G., � physik. Chem., A 159 , 27 (1932). Rulfs, C. L. and Meinke, W. w., l.:_ Am. Chem . Soc., 74, 235 (1952). 126

Selwood, P. w., 11Magnetochemistry11 , Interscience, New York, N. Y., 1943, p. 203. Smith, W. T., Jr., Line, L. and Bell, w. A., .!!.:_ Am. Chem. Soc., in press (1952) . Smith, w. T., Jr., Private Comrmmication, 1951. Von Hevesy, G., Chem. �, ], 321 (1927). Wadsley, A. n. and Walkley, H., !!.:_ Electrochem. Soc., 95, 11 (1949). Washburn, E. W., !!_:. Research Natl. Bur. Standards, 10, 525 (1933). 11 White, W. P., "The Modern Calorimeter , The Chemical Catalog Company, New York, N. Y., 1928, p. 169. Zachariasen, w. H., Argonne National Laboratory Report, ANL-4082, 1947; to be published in Acta. Cryst. Zachariasen, w. H. , Oak Ridge National Laboratory Memorandum, CF-47-11-464 (1947). VITA

James w. Cobble was born in Kansas City, Missouri, on March 15, 1926. He attended the elementary schools of this city.andwas graduated from San Diego High School in San Diego, California in June, 1942 . The following two years he at tended San Diego State College, leaving to join the US Naval V-12 unit at Arizona State College at Flagstaff, Arizona in March, 1944. In March, 1945, he entered the US

I Naval Midshipman s School at the University of Notre Dame, and was commissioned an Ensign, USNR in July, 1945. He served with the Pacific Fleet until May, 1946, receiving his B. A. degree from Arizona State (in absentia) in February, 1946. He attended the University of California at Berkeley, California from September, 1946 until September, 1947 . He entered the Graduate School of the University of Southern California at Los Angeles, California in September, 1947 and received the M.S. degree in Chemistry from that institution in June, 1949. During his tenure at that school, he held successively the positions of teaching assistant, research assistant, a..l'ld research fellow. He has been employed as a chemist at the Oak Ridge National Laboratory at Oak Ridge, Tennessee from January, 1950 to the present time. In June, 1950 he entered the Graduate School of the University of Tennessee as an Oak Ridge Resident Student, and was admitted to candidacy for the Ph. n� degree in January, 1952. The author is a member of the American Chemical Society, Sigma Xi, Phi Kappa Phi, and Phi Lambda Upsilon, and has published the following papers : "The Precision Determination· of Some Half-lives", J. w. Cobble and R. w. Atteberry, Phys . Rev. , 80, 917 (19.50). "The Electron Transfer Exchange Reaction of Ferricyanide and Ferrocyanide Ions", J. w. Cobble and A. W. Adamson, ---J. Am. Chern. Soc., 72, 2276 (19.50).

"Recoil Reactions with High Intensity Slow Neutron Sources I. The Szilard-Chalmers Enrichment of 3.5 �9 h Br8211 , G. E. Boyd, J. Cobble, and Sol Wexler , � Am. Chern. Soc. , 74, 237 (19.52). w.

"Recoil Reactions with High Intensity Slow Neutron Sources II . The Retention of Radiobromine by Crystalline KBr03", J. w. Cobble and G. E. };loyd, J. Arn. Chern. Soc., 74, 1282 (19.52 ). "Chemistry of Technetium I. Preparation of Technetium Heptoxide" , G. E. Boyd, J. Cobble, M. Nelson and T. Smith, Jr., -J. --Am. --Chem. · Soc., 74, 556 (1w.9.52 ). c. w.

"Chemistry of Technetium II. Preparation of Technetium Metal", J. Cobble , M. Nelson, G. Parker, T. Smith, Jr., and G. E. Boydw., !!.:_ Am. Chem.c. Soc., 74, 1852w. (1952) . w. "Differential Diffusion Coefficients of Sodium Ion in Sodium Chloride Solut�ons" , A. Adamson, J. Cobble, and J. M. Nielsen, {:_� Phys ., 17, 740 w.(1 949). w.

11 The Self Diffusion Coefficients of the Ions in Aqueous Sodium ° Chloride and Sodium Sulfate at 25 11 ; J. M. Nielsen, A. W. Adamson, ani J. W. Cobble, i_: Am. Chern. Soc., 74, 446 (19.52). "The Thermodynamics of Technetium arrl Its Compounds I. The Vapor Pressures of Technetium Heptoxide , Pertechnic Acid arrl Its Saturated Solution", T. Smith, Jr. , G. E. Boyd, and J. Cobble, ---J. Am. Chem. Soc., (in prw.ess) (19.52). w. "Calibration of the 477' High Pressure ·Ion Chamber" , J. w. Cobble, Oak Ridge National Laboratory Report, CF-50-6-114 (19.50).

"Experiment s in Reactor Physics : Experimr nt 7, The Ion Exchange Separation of Cobalt and Nickel", J. W. Cobble , Oak Ridge Reactor School (19.51).